Euphorbiaceae), with Emphasis on Two Species Of

UNIVERSIDADE DE BRASÍLIA INSTITUTO DE CIÊNCIAS BIOLÓGICAS

HANNAH CASCELLI FARINASSO

FLORAL BIOLOGY AND POLLINATION OF MANIHOT MILL.

(EUPHORBIACEAE), WITH EMPHASIS ON TWO SPECIES OF

THE CERRADO

BIOLOGIA FLORAL E POLINIZAÇÃO DE MANIHOT MILL.

(EUPHORBIACEAE), COM ÊNFASE EM DUAS ESPÉCIES DO

CERRADO

Brasília

2019

HANNAH CASCELLI FARINASSO

FLORAL BIOLOGY AND POLLINATION OF MANIHOT

MILL. (EUPHORBIACEAE), WITH EMPHASIS ON TWO

SPECIES OF THE CERRADO

BIOLOGIA FLORAL E POLINIZAÇÃO DE MANIHOT MILL.

(EUPHORBIACEAE), COM ÊNFASE EM DUAS ESPÉCIES

DO CERRADO

Dissertação apresentada ao Instituto de Ciências Biológicas da Universidade de Brasília (UnB) como

parte dos requisitos exigidos para obtenção do Título de Mestra em Botânica

Orientador: Prof. Dr. Hélder Nagai Consolaro

Coorientador: Prof. Dr. Antonio José Camillo de Aguiar

Brasília

2019

DEDICATÓRIA

Ao Cerrado, por sua imensa diversidade de formas, que abriga tamanha

fonte de riqueza e inspiração.

À rainha do Brasil, a mandioca e ao seu gênero, por ser tão interessante e

apaixonante, e a todos os antepassados indígenas que domesticaram e

criaram esse tesouro.

AGRADECIMENTOS

Aos orientadores Hélder e Antonio, pela disponibilidade, paciência e aprendizado. Aos colaboradores

Moises e Sueli, fundamentais na dissertação. A todos os membros das bancas, que contribuíram para o melhor desenvolvimento do trabalho. Aos pesquisadores doutores que identificaram espécimes entomológicos: José

Pujol, Julia Calhau, Paula Riccardi, Rosaly Rocha, Silvio Nihei (Diptera); Rodrigo Feitosa (Formicidae); Luís

Lira (Coleoptera). Sou grata a Karla Monique Valadão, imprescindível para a inclusão da anatomia e histoquímica das glândulas de M. esculenta no estudo.

Ao Jaime Sautchuk, pelo consentimento para a realização da pesquisa com a espécie M. oligantha, na

RPPN Serra Linda dos Topázios. Também sou grata ao Ibram pela concessão e permissão de coleta para o estudo dentro do Parque Dom Bosco. As seguintes instituições e seus técnicos também foram de suma importância para a realização da dissertação: o laboratório de Microscopia e Microanálise do Instituto de

Ciências Biológicas da UnB; a Embrapa Cenargen, e a sua Coleção viva do gênero Manihot, e o Laboratório de Microscopia; a Embrapa Cerrados; a Estação Ecológica da UnB.

Aos que auxiliaram em campo: Vítor, Raphael, Gabriela, Raquel, Caracol, Bebela, Gustavo, Carol,

Bruno, os Matheus, Mayco, Moises, Wagner, Rogério, e Antonio que também fotografou o comportamento dos polinizadores nas flores de Manihot. A todos os professores, colegas das disciplinas e os companheiros dos laboratórios de Hymenoptera e Biologia Reprodutiva de Angiospermas.

Agradeço ao Vítor, maravilhoso, essencial em diversos momentos da dissertação, desde trabalho e incentivo moral no campo a revisões de manuscritos. Aos ciclos de familiares e amizades por apoiarem meus estudos e escolhas.

À Universidade de Brasília e ao PPG-Botânica.

À CAPES pela bolsa de mestrado.

RESUMO

As espécies do gênero Manihot, das quais M. esculenta apresenta importância alimentícia, são pouco conhecidas sobre sua biologia reprodutiva. Nesta dissertação investigamos caracteres florais, reprodutivos e interações planta-polinizador em Manihot, focando em espécies do Cerrado, bioma com área prioritária para conservação do gênero. No primeiro capítulo, comparamos espécies de distribuição restrita no Cerrado a espécies mais amplamente distribuídas. Revelamos a ocorrência de osmóforos, glândulas que emitem odores, nas flores de todas as 16 espécies analisadas. As estruturas das glândulas florais mantiveram-se conservadas e aplicando testes histoquímicos em três espécies, identificamos compostos que potencialmente contribuem na proteção das flores e atração de polinizadores (terpenos). As espécies endêmicas do Cerrado se diferenciaram no design da flor feminina e morfologia dos osmóforos, indicando maior diversificação nesse bioma. No segundo capítulo, demonstramos em duas espécies de Manihot do Cerrado, sistemas reprodutivos que dependem de polinizadores para a formação de sementes. Variados grupos funcionais de insetos atuaram como polinizadores, porém as espécies podem manifestar características florais, como baixa concentração do néctar, que resultam na restrição de determinados visitantes. As duas espécies diferem nos principais polinizadores (abelhas versus formigas e moscas) e, independente do sistema de polinização generalista, evidenciamos filtros florais. Ademais, a monoicia limita possibilidades de visitantes pouco frequentes polinizarem pelo baixo número de flores femininas. Concluindo, apesar de possuir características florais conservadas, as espécies de Manihot podem diferir em sua biologia da polinização, tendendo a depender de grupos de insetos mais abundantes para seu sucesso reprodutivo.

Palavras-chave biologia reprodutiva da mandioca, endemismo, flores unissexuais, interações planta- inseto, metabólitos secundários, parentes silvestres de culturas

ABSTRACT

The species of the genus Manihot, whose M. esculenta is of nutritional importance, are poorly known about their reproductive biology. In this dissertation we investigated floral and reproductive characters and plant-pollination interactions in Manihot, focusing on species of the Cerrado, a biome with priority area for conservation of the genus. In the first chapter, we compared species with restricted distribution in the Cerrado to more widely distributed species. We revealed the occurrence of osmophores, scent glands, in the flowers of all 16 species analyzed. The structures of the floral glands were conserved and applying histochemical tests in three species, we identified compounds that potentially contribute to the protection of the flowers and attraction of pollinators (terpenes). The endemic species of the Cerrado differed in the design of the female flower and morphology of osmophores, indicating greater diversification in this biome. In the second chapter, we demonstrated in two Cerrado Manihot species, reproductive systems that depend on pollinators for seed formation.

Various functional groups of insects acted as pollinators, but the species can manifest floral characteristics, such as low concentration of nectar, resulting in the restriction of certain visitors. The two species differ in the main pollinators (bees versus ants and flies) and, regardless of the generalist pollination system, we evidenced floral filters. Additionally, monoecy limits the possibilities of infrequent visitors to pollinate due to the low number of female flowers. In conclusion, despite having conservative floral traits, Manihot species can differ in their pollination biology, tending to depend on more abundant groups of insects for reproductive success.

Keywords crop wild relatives, endemism, plant-insect interactions, reproductive biology of cassava, secondary metabolites, unisexual flowers

SUMÁRIO

GENERAL INTRODUCTION ...... 8 REFERENCES ...... 10 CHAPTER 1 Optimizing gland functions: defense and attraction evidence in conservative floral structures of Manihot species (Euphorbiaceae)...... 14 ABSTRACT ...... 15 INTRODUCTION ...... 16 MATERIALS AND METHODS ...... 18 RESULTS ...... 21 DISCUSSION ...... 35 REFERENCES ...... 41 CHAPTER 2 Ecological generalized but not so much: pollinator filtering in monoecious Manihot (Euphorbiaceae) ...... 53 ABSTRACT ...... 54 INTRODUCTION ...... 55 MATERIAL AND METHODS ...... 57 RESULTS ...... 61 DISCUSSION ...... 71 REFERENCES ...... 77 APPENDIX ...... 87

GENERAL INTRODUCTION

Manihot is a neotropical genus with recent diversification in South America, reaching the highest diversity and endemism in Brazilian Cerrado (Rogers and Appan 1973, Chacón et al. 2008,

Duputié et al. 2011). The genus is recognized worldwide due to cassava (M. esculenta), whose tuberous root is an important food source of carbohydrate for over 700 million people (http:// faostat.fao.org). Cassava has a great potential for improvement by gene pool from the wild relatives of the species (Nassar and Hashimoto 2006, Nassar et al. 2008, 2012, Castañeda-Álvarez et al.

2016). However, many Manihot species are endangered (IUCN 1997). Furthermore, from Cerrado a high proportion (41%) of endemic species are threatened, mainly due to habitat loss (Simon et al.

2018).

Manihot presents unisexual flowers (Rogers and Appan 1973), which is uncommon amongst flowering plants (Tanurdzic and Banks 2004). The genus features well-conserved floral structure, with flowers in a single whorl formed by five tepals and have on the bottom of the flower a nectariferous gland (Rogers and Appan 1973). The flowers are insect pollinated, mainly by bees and wasps (Rogers and Appan 1973, Nassar and Carvalho 1990). Manihot flowers are often dish to bowl shape that is thought to be generalist pollinated (Olesen et al. 2007). Although, studies in natural conditions are lacking for the genus and interpretations about pollination systems based solely on floral morphology can be misleading (Lindsey and Bell 1985, Ollerton et al. 2007). Consequently, here, we should look at field conditions to understand plant-pollinator interactions.

According to Rogers and Appan (1973), the sexual system of the genus is predominantly monoecious (individuals bearing both unisexual flowers), with a minority of dioecious species

(unisexual individuals). This is consistent with what is known for the Euphorbiaceae, whereby the monoecious system is predominant (Webster 1994, Bullock 1985, Machado et al. 2006, Chen and Li

2008). Moreover, monoecy can be associated with protogyny (Bertin and Newman 1993), indeed

Manihot species are protogynous at inflorescence level (Halsey et al. 2008, Perera et al. 2013): female flowers function prior to male flowers, therefore the sexes are temporally separated

(dichogamy). The dichogamy evolves in the presence of sex interference and inbreeds depression, though the match between the time of pollen dispersal and available ovules commonly results in asymmetric dichogamy (Sargent et al. 2006). Despite that, the protogyny evolution act minimizing the sex interference, increases the crossing rate and potentially avoid self-fertilization (Bertin 1993,

Barrett 2003, Narbona et al. 2011, Pang et al. 2013). These premises can be applied to Manihot species, in which the breeding system is out-crosser (McKey et al. 2005), but sometimes male and female flowers are open at the same time and geitonogamy becomes possible (da Silva et al. 2003,

Lai et al. 2014).

In flowers with separated sexes (space or time), different floral traits might evolve (Hemborg and Bond 2005, Barrett and Hough 2013). Flower size dimorphism tends to be predominant in taxa bearing unisexual flowers, whereby 85% revealed sexual dimorphism in size; in addition, the relative size was suggested to be influenced by the climate zone: in the temperate zone the male flower is often larger, and in tropics zones both sexes display equally odds to be larger (Delph et al. 1996).

Flower size dimorphism is also conspicuous to endemic taxa (Humeau et al. 2003). In Manihot,

Byrne (1984) reported that M. esculenta female flowers are twice the size of male flowers. Although in Cerrado species the male flowers bigger than females seems prevalent (Silva and Sodré 2014,

Mendoza et al. 2015, 2016). Additionally, Manihot present sexually dimorphic perianth, with the female flowers dish-shaped and on the other way, the male flowers are generally bell-shaped. Plants bearing unisexual flowers allow comprehension of selection pressures on each sexual component

(Costich and Meagher 2001, Rocheta et al. 2014). However, shape dimorphism still appears a poorly explored subject (Miller and Venable 2003). The evolutionary pressures resulting in the dimorphic shape of Manihot unisexual flowers could represent a matter of interest.

Considering the unexplored fields in the reproductive biology of Manihot and its main center of endemism in Cerrado, we proposed different perspectives in this study. In the first chapter, the focus is on the Manihot flowers, in which we investigated two floral glands known for mediate interactions with pollinators – osmophores and nectaries – providing new information for the genus.

Besides, we reviewed floral features and support the idea that Cerrado endemic species manifested diversified traits among the genus. The second chapter presents the pollination and breeding systems of two wild Manihot species of Cerrado. In addition, we evaluated floral biology in male and female flowers giving insight into pollinator filters functioning in a monoecious plant.

REFERENCES

Barrett, S. C. H. (2003). Mating strategies in flowering plants: the outcrossing-selfing paradigm and beyond. Philosophical Transactions of the Royal Society B: Biological Sciences, 358(1434), 991–1004. https://doi.org/10.1098/rstb.2003.1301

Barrett, S. C. H., & Hough, J. (2013). Sexual dimorphism in flowering plants. Journal of Experimental Botany, 63(2), 695–709. https://doi.org/10.1093/jxb/err313

Bertin, R. I. (1993). Incidence of monoecy and dichogamy in relation to self-fertilization in angiosperms. American Journal of Botany, 80(5), 557–560. https://doi.org/10.2307/2445372

Bertin, R. I., & Newman, C. M. (1993). Dichogamy in angiosperms. The Botanical Review, 59(2), 112–152. https://doi.org/10.1007/BF02856676

Byrne, D. (1984). Breeding Cassava. Chapter 3 in: Plant Breeding Reviews, Vol. 2.

Castañeda-Álvarez, N. P., Khoury, C. K., Achicanoy, H. A., Bernau, V., Dempewolf, H., Eastwood, R. J., Toll, J. (2016). Global conservation priorities for crop wild relatives. Nature Plants, 2(4), 1–6. https://doi.org/10.1038/NPLANTS.2016.22

Chacón, J., Madriñán, S., Debouck, D., Rodríguez, F., & Tohme, J. (2008). Phylogenetic patterns in the genus Manihot (Euphorbiaceae) inferred from a combined analysis of six DNA regions. Molecular Phylogenetics and Evolution, 49, 260–267.

Chen, X. S., & Li, Q. J. (2008). Sexual systems and ecological correlates in an azonal tropical forests, SW China. Biotropica, 40(2), 160–167. https://doi.org/10.1111/j.1744- 7429.2007.00364.x

Costich, D. E., & Meagher, T. R. (2001). Impacts of floral gender and whole-plant gender on floral evolution in Ecballium elaterium (Cucurbitaceae). Biological Journal of the Linnean Society, 74(4), 475–487. https://doi.org/10.1006/bijl.2001.0600

Da Silva, R. M., Bandel, G., & Martins, P. S. (2003). Mating system in an experimental garden composed of cassava (Manihot esculenta Crantz) ethnovarieties. Euphytica, 134(2), 127–135. https://doi.org/10.1023/B:EUPH.0000003644.60126.4a

Delph, L. F., Galloway, L. F., & Stanton, M. L. (1996). Sexual dimorphism in flower size. The American Naturalist, 148(2), 299–320. https://doi.org/10.1086/285926

Duputié, A., Salick, J., & Mckey, D. (2011). Evolutionary biogeography of Manihot (Euphorbiaceae), a rapidly radiating Neotropical genus restricted to dry environments. Journal of Biogeography, 1:1-11. https://doi.org/10.1111/j.1365-2699.2011.02474.x

Halsey, M. E., Olsen, K. M., Taylor, N. J., & Chavarriaga-Aguirre, P. (2008). Reproductive Biology of Cassava (Manihot esculenta Crantz) and Isolation of Experimental Field Trials. Crop Science, 48, 49–58. https://doi.org/10.2135/cropsci2007.05.0279

Hemborg, Å. M., & Bond, W. J. (2005). Different rewards in female and male flowers can explain the evolution of sexual dimorphism in plants. Biological Journal of the Linnean Society, 85(1), 97–109. https://doi.org/10.1111/j.1095-8312.2005.00477.x

Humeau, L., Pailler, T., & Thompson, J. D. (2003). Flower size dimorphism in diclinous plants native to La Réunion Island. Plant Systematics and Evolution, 240(1–4), 163–173. https://doi.org/10.1007/s00606-003-0011-4

IUCN. (1997). Red List of Threatened Plants. Magnoliopsida. Compiled by the World Conservation Monitoring Center. IUCN - The World Conservation Union, Gland, Switzerland and Cambridge, UK. 862pp.

Lai, H., Chen, X., Chen, Z., Zhou, Y., Ou, W., Li, K., Chen, S. (2014). In situ Pollen Germination and Artificial Pollination Compatibility in Cassava (Manihot esculenta Crantz). Applied Mechanics and Materials, 651–653, 245–251. https://doi.org/10.4028/www.scientific.net/AMM.651-653.245

Lindsey, A. H., & Bell, R. (1985). Reproductive biology of Apiaceae. II. Cryptic specialization and floral evolution. American Journal of Botany, 72(2), 231–247. https://doi.org/10.1002/j.1537- 2197.1985.tb08288.x

Machado, I. C., Lopes, A. V., & Sazima, M. (2006). Plant sexual systems and a review of the breeding system studies in the Caatinga, a Brazilian tropical dry forest. Annals of Botany, 97(2), 277–287. https://doi.org/10.1093/aob/mcj029

McKey, D., Elias, M., Pujol, B., & Duputié, A. (2005). The evolutionary ecology of clonally propagated domesticated plants. New Phytologist, 186, 318–332. https://doi.org/10.1111/j.1469- 8137.2010.03210.x Key

Mendoza F, J. M., Simon, M. F., Reis, T. S., Fidelis, A., & Cavalcanti, B. (2016). New endemic species of Manihot (Euphorbiaceae) from Serra do Tombador in Goiás, Central Brazil. Phytotaxa, 273(3), 147–157. https://doi.org/10.11646/phytotaxa.273.3.1

Mendoza, M., Simon, M. F., & Cavalcanti, T. B. (2015). Three new endemic species of Manihot (Euphorbiaceae) from the Chapada dos Veadeiros, Brazil. Arnaldoa, 22(2), 297–312.

Miller, J. S., & Venable, D. L. (2003). Floral Morphometrics and the Evolution of Sexual Dimorphism in Lycium (Solanaceae). Evolution, 57(1), 74–86.

Narbona, E., Ortiz, P. L., & Arista, M. (2011). Synchrony in two Euphorbia species: alternative mechanisms for avoiding self-fertilization? PLoS ONE, 6(6). https://doi.org/10.1371/journal.pone.0020668

Nassar, N. M. A., & Carvalho, C. G. P. (1990). Insetos polinizadores e seus comportamentos nas espécies silvestres da mandioca, Manihot spp. Journal of Brazilian Association for Advancement of Science, 42, 703–705.

Nassar, N. M. A., & Hashimoto, D. Y. C. (2006). Wild cassava, Manihot spp.: A survey of useful species with reference to economic value. Gene Conserve, 5(19), 284:295.

Nassar, N. M. A., Hashimoto, D. Y. C., & Fernandes, S. D. C. (2008). Wild Manihot species: botanical aspects , geographic distribution and economic value. Genetics and Molecular Research, 7(1), 16–28.

Nassar, N. M. A., Mendoza, F. J. M., & Sano, N. (2012). Some interesting cassava cultivars: 9 – ICB 300. Geneconserve, 11(43), 22–25.

Olesen, J. M., Dupont, Y. L., Ehlers, B. K., & Hansen, D. M. (2007). The openness of a flower and its number of flower-visitor species. Taxon, 56(3), 729–736. https://doi.org/10.2307/25065857

Ollerton, J., Killick, A., Lamborn, E., Watts, S., & Whiston, M. (2007). Multiple meanings and modes: on the many ways to be a generalist flower. Taxon, 56(3), 717–728. https://doi.org/10.2307/25065856

Pang, C. C., Scharaschkin, T., Su, Y. C. F., & Saunders, R. M. K. (2013). Functional monoecy due to delayed anther dehiscence: a novel mechanism in Pseuduvaria mulgraveana (Annonaceae). PLoS ONE, 8(3). https://doi.org/10.1371/journal.pone.0059951

Perera, P. I. P., Quintero, M., Dedicova, B., Kularatne, J. D. J. S., & Ceballos, H. (2013). Comparative morphology, biology and histology of reproductive development in three lines of Manihot esculenta Crantz (Euphorbiaceae: Crotonoideae). AoB PLANTS, 1–18. https://doi.org/10.1093/aobpla/pls046

Rocheta, M., Sobral, R., Magalhães, J., Amorim, M. I., Ribeiro, T., Pinheiro, M., Costa, M. M. R. (2014). Comparative transcriptomic analysis of male and female flowers of monoecious Quercus suber. Frontiers in Plant Science, 5, 1–16. https://doi.org/10.3389/fpls.2014.00599

Rogers, D. J., & Appan, S. G. (1973). Manihot Manihotoides (Euphorbiaceae). Flora Neotropica, 13, 1–272.

Sargent, R. D., Mandegar, M. a, & Otto, S. P. (2006). A model of the evolution of dichogamy incorporating sex-ratio selection, anther-stigma interference, and inbreeding depression. Evolution, 60(5), 934–944. https://doi.org/10.1111/j.0014-3820.2006.tb01172.x

Silva, M. J., & Sodré, R. C. (2014). A dwarf species of Manihot Mill. (Euphorbiaceae s. s.) from the highlands of Goiás, Brazil. Systematic Botany, 39(1), 222–226. https://doi.org/10.1600/036364414X678134

Simon, M. F., Reis, T. S., Mendoza F., J. M., Arquelão, T. K. M., Bringel, J. B. A., Noronha, S. E., Scariot, A. (2018). Conservation assessment of cassava wild relatives in central Brazil. Biodiversity and Conservation, (0123456789), 1–24. https://doi.org/10.1007/s10531-018-1626- 7

Tanurdzic, M., & Banks, J. A. (2004). Sex-determining mechanisms in land plants. The Plant Cell, 16. https://doi.org/10.1105/tpc.016667

Webster, G. L. (1994). Classification of the Euphorbiaceae. Annals of the Missouri Botanical Garden, 81(1), 3–32.

CHAPTER 1

Optimizing gland functions: defense and attraction evidence in conservative

floral structures of Manihot species (Euphorbiaceae)

Hannah C. Farinasso¹, Antonio J. C. Aguiar², J. Moises Mendoza F.³, Hélder

Consolaro4, Sueli M. Gomes¹

¹Departamento de Botânica, Instituto de Ciências Biológicas, Universidade de Brasília, DF, Brasil.

²Departamento de Zoologia, Instituto de Ciências Biológicas, Universidade de Brasília, DF, Brasil.

³Herbario del Oriente Boliviano, Museo de Historia Natural Noel Kempff Mercado, Santa Cruz, Bolivia.

4Instituto de Biotecnologia, Universidade Federal de Goiás, Regional Catalão, GO, Brasil.

Manuscript to be submitted to the Journal Plant Systematics and Evolution

ABSTRACT

Floral traits evolution undergoes selective pressures by antagonists and mutualists, i.e. secretory structures and glands function for defense and pollinators attraction. We explored floral features and glands of Manihot, to give insights into the genus floral evolution. Furthermore, we evaluated if narrowly distributed species of the main diversity center of the genus – Cerrado – present differential floral traits. We studied 16 species, eight narrowly distributed in Cerrado and eight more widely distributed or from a different biome. We reviewed in literature the species floral traits, examined its flowers and performed red neutral test to detect secretory activity. In three species, we analyzed the structure and histochemistry of the glands. Our data supported two floral glands in Manihot, nectary, and osmophores (scent gland), providing the second register of osmophores in Euphorbiaceae.

Narrow endemic species of Cerrado had differential floral features in female flower morphology and size and more often had deciduous bracts. Species endemic to Cerrado also appeared with more conspicuous osmophores. Our findings indicate that the Cerrado species evolved differential floral traits in Manihot. Although species presented individual variation in floral traits, a very conservative state for both floral glands evinced. Both glands had secretory structures and compounds related to defense and pollinator attraction. Due to osmophore pattern in tepals and its main chemicals, it seems that this gland evolved first to function in floral protection. In summary, a double role of floral glands in attraction and defense should be common, especially in osmophores, thereby optimizing plant reproductive fitness and defense system.

Keywords alkaloids, cassava, endemism, monochlamydeous, monoecy, scanning electron microscopy, terpenes, unisexual flowers

INTRODUCTION

Floral glands have multiple independent origins and its secretion is known to mediate complex interactions of plant-pollinators since the early evolution of angiosperms (Pacini et al. 2003,

Bernardello 2007, Nepi et al. 2009, Erbar 2014, Poinar and Poinar 2018). Nectar is probably the most common floral reward and it can be an innovation that increases reproductive success among animal pollinated taxa (Simpson and Neff, 1981, Neiland and Wilcock 1998, Tong et al. 2018). In pollination, besides rewarding glands, there are also scent glands (or osmophores) that act in plant- pollinator communication by floral volatile emission (Vogel 1990, Piechulla and Pott 2003, Raguso

2008). A variety of tissues can emit floral scents, in specialized areas or in a diffuse way, occurring in different organs, such as petals, sepals, anther connective, and receptacle (Vogel 1962, Sazima et al. 1993, Vogel and Hadacek 2004, Bergougnoux et al. 2007, Kowalkowska et al. 2014, Possobom et al. 2015, Wiśniewska et al. 2018). The importance of scent emission in floral evolution is yet to be highlighted. In fact, odor signals in floral recognition, mainly for insect pollinators, can constitute an accurate and honest signal of rewards, even greater than flower size and color (Knudsen et al. 2001,

Wright and Schiestl 2009, Parachnowitsch et al. 2012). Moreover, higher emission of floral scents is related to a higher reproductive fitness of the plants (Majetic et al. 2009).

In addition to floral glands acting in pollinator rewarding and attraction, defense function is found for both scent glands and nectaries (Domínguez et al. 1989, Thornburg et al. 2003, Narbona and Dirzo 2010, Marques et al. 2015, Płachno et al. 2017). Compounds of floral scents and nectar can simultaneously attract pollinators and defend against antagonists (Junker and Bluthgen 2010,

Junker et al. 2011, Huang et al. 2012). The dual issues of plant reproduction and defense have been shown recurrent in floral traits (Strauss 1997, Pichersky and Gershenzon 2002, Irwin et al. 2004, Sun et al. 2008, Kessler et al. 2008, Kessler and Halitschke 2009, Johnson et al. 2015, Sletvold et al.

2015, Veiga et al. 2015, Lucas-Barbosa 2016), despite that works about only one gland function

17 predominate in the literature (e.g. Fahn 2000, Machado et al. 2015, Demarco 2017, Płachno et al.

2018).

The family Euphorbiaceae for example, possesses both secretory structures related to defense such as, laticifers, colleters, idioblasts, trichomes, and structures for pollinator attraction such as nectaries, resin reward, and osmophores (Armbruster 1984, Freitas et al 2001, Machado et al. 2015,

Vitarelli et al. 2015, Gagliardi et al. 2016). The family is noteworthy by the diversity of secondary compounds (Solereder 1908, Metcalfe and Chalk 1950, Thomas 1991, Seigle 1994). Another remarkable feature is the modified inflorescences (cyathium) that are highly specialized (Prenner and

Rudall 2007, Thakur and Patil 2011), even though restricted to the tribe Euphorbieae (Steinmann and

Porter 2002). Nevertheless, in the floral evolution of Euphorbiaceae, reductions in number and types of whorls are frequent (Venkata-Rao and Ramalakshmi 1968, Webster 1994, De-Paula et al. 2011).

Simultaneously to floral morphology evolution, the glands also diversified within euphorbiaceous lineages (Armbruster 1985, Cacho et al. 2010).

Manihot is a monophyletic genus of Euphorbiaceae from Neotropical region that probably originates in Mesoamerica, and recently (ca 3.5 Ma) radiated in South America (Chacón et al. 2008,

Duputié et al. 2011). Manihot esculenta, the cassava, is worldwide recognized as an important crop for human food supplier (Hershey 2010). It is known that the mating system of M. esculenta is predominantly outcrossing and flowers are insect pollinated (Rogers and Appan 1973, Nassar and

Carvalho 1990, da-Silva et al. 2003, Lai et al. 2014). The related wild species are a potential resource for the genetic improvement of commercial species (Nassar et al. 2012, An et al. 2016). The main diversity and endemism of Manihot are in the Cerrado of central Brazil, with 75 species

(Rogers and Appan 1973, Duputié et al. 2011, Simon et al. 2018). The Cerrado is the highest priority area for conservation of cassava wild relatives (Castañeda-Álvarez et al. 2016) and many of its endemic species are currently threatened (Simon et al. 2018). It was revealed that many Cerrado lineages are derived and recent, which diversified in situ and adapted to frequent fires (Simon et al.

2009). In fact, the Manihot appeared to rapidly evolve in this biome (Duputié et al. 2011). A study with narrow endemic species supports differential reproductive traits comparing to its wide-spread congeners (Lavergne et al. 2004). Therefore, we could expect differential floral traits in endemic

Manihot species of Cerrado.

Despite that, the genus presents a well-conservative floral structure, with unisexual flowers reduced in a single whorl formed by five tepals (Rogers and Appan 1973, Webster 1994, Carmo

Júnior et al. 2013). Certainly, some traits appear constant in Manihot flowers as the nectary named basal or floral disk (Venkata-Rao and Ramalakshmi 1968, Rogers and Appan 1973, Perera et al.

2013); though the nature of secretion or gland structure was not reported. In addition, we recognized a modified region in tepal margins, yellow glandlike, which was not studied so far. Overall, floral features of the genus remain untapped. Therefore, we aimed to explore the following questions in

Manihot floral evolution: (1) how conservative are the floral traits of Manihot in South America, more specifically, in Cerrado? Do the endemic species with a narrow distribution in Cerrado present different floral traits? (2) Which are the floral glands and secretory structures in Manihot and how conservative are they? What are the possible floral gland functions?

MATERIALS AND METHODS

Studied species

We studied 16 species of Manihot belonging to four of the seven subclades of Manihot phylogeny, mainly from dry climates (Duputié 2011). Most investigated taxa are endemic to

Cerrado, as following: M. congesta M. Mend. & T.B. Cavalc., M. debilis M. Mend. & T.B. Cavalc.,

M. fruticulosa (Pax) Rogers & Appan, M. minima M. Mend. & T.B. Cavalc., M. pinatiloba M.

Mend. & T.B. Cavalc., M. porphyrantha M. Mend. & T.B. Cavalc., M. purpurea M. Mend. & T.B.

Cavalc., M. robusta M. Mend. & T.B. Cavalc. and M. sagittata M. Mend. & M. Martins. The material of these taxa was obtained from the living collection of the Brazilian Agricultural Research

Corporation (Embrapa)/Genetic Resources and Biotechnology (Cenargen) collected originally from in-situ populations under collection numbers of M. Mendoza (see Table 1 for voucher details). The

M. esculenta was included in our samplings and assessed from a cultivar from Embrapa Cerrados.

Additionally, M. glaziovii Müell. Arg. and M. pseudoglaziovii Pax & K. Hoffm. from Caatinga biome were obtained from the living collection of the Estação Biológica da Universidade de Brasília.

Other species were collected in their natural habitat in Cerrado: M. gracilis Pohl, M. oligantha Pax,

M. tripartita (Spreng.) Müll.Arg. and M. violacea Pohl.

Floral traits in Manihot and species distribution range

We reviewed from the literature the floral traits and biome for the analyzed species (Rogers and Appan 1973, Byrne 1984, Mendoza et al. 2015, 2016, 2018, Mendoza and Martins 2018). Data of floral anthesis, nectar traits and pollinators are largely unknown for the wild species. Therefore, the assessed information was according to its availability for most species, which consisted of bracts features, male and female flower shape, length size (mean of the maximum and minimum measurements) and flower color.

We used the distribution range provided by the recent study of Simon et al. (2018), in which we established an endemism criterion based on the extent of occurrence (EOO) of the species.

Defining narrow endemism can be very difficult and several factors should be considered

(Kruckeberg and Rabinowitz 1985), thus it might explain why there is not an absolute definition

(Médail and Baumel 2018). Here, we considered that species with EOO smaller than 170 km² as narrow distributed (see Pisanu et al. 2009).

Gland morphology and secretory activity

To assess floral glands, its morphology, the presence of secretion and possible differences between sexes, we collected fresh flowers from the studied species and examined under a stereoscopic microscope (Leica M125). Plant material was collected according to its availability, but often we were limited by the low number of available flowers in anthesis. Despite that at least six intact flowers from each species were submerged in neutral red to detect secretory activity (Vogel

1990) and photographed before and after the test (AmScope FMA050).

Gland structure in scanning electron microscopy

For micromorphology analyses, we carried out scanning electron microscopy (SEM) in the floral location with secretory activity. We selected three species, M. esculenta, M. violacea, and M. oligantha, based on the availability of flowers. Open fresh flowers were fixed in formaldehyde- glutaraldehyde 2.5% for 24h (Karnovsky 1965) and conditioned in phosphate buffer 0.1 M, pH 7.3.

The samples were post-fixed with osmium tetroxide 1% for 30 min, dehydrated in acetone series, dried at the critical point using liquid CO2 (Balzers, CPD 030), mounted on stubs, coated with a golden layer (Balzers, SCD 050) and analyzed in SEM (Jeol JSM-7000F).

Gland structure in optical microscopy

For inner secretory structure analysis, we collected male and female flowers from the same species submitted to SEM. We sampled recent open flowers and fixed in neutral buffered formalin during 48h (Lillie 1965), then embedded in histological paraffin (Johansen 1940). The longitudinal and transversal sections 8-10 µm thick were obtained in a rotary microtome and stained with safranin and toluidine blue (Johansen 1940). The slides were mounted in synthetic resin (Paiva et al. 2006).

Additionally, we applied a different protocol for minute regions on tepals with secretory activity, for the species M. esculenta and M. violacea. Initially, the procedure of fixation and post-fixation followed as described for SEM. Afterward, we added uranyl acetate for contrast and embedded the

21 samplings in epoxy resin (Araldite). Semi-thin sections with 70-80 nm thick were obtained using ultra-microtome and stained with toluidine blue 1%.

Histochemical analysis

For histochemical analyses of the floral glands, we prepared the samples following the protocol described above for paraffin (Johansen 1940), with exception of the reagents for alkaloids and terpenes detection that we used fresh samples, sectioned by hand-cut. We employed histochemical tests for the main compounds classes detection: 1) lugol (iodine and potassium iodide) for starch (Johansen 1940); 2) periodic acid/Schiff reagent for polysaccharides (McManus 1948); 3)

Comassie R250 for protein (Fisher 1968); 4) ruthenium red for acid mucilage (Gregory and Baas

1989); 5) Sudan IV and Sudan Black B for total lipids (Pearse 1985); 6) ferric chloride III for phenolic compounds (Johansen 1940); 7) Wagner’s reagent for alkaloids (Furr and Mahlberg 1981);

8) NADI for terpenoids (David and Carde 1964). The control and slide mounting followed the procedure of the respective test. The sections were analyzed and the results were registered in a photomicroscope Olympus, U-TV0.5xC-3 with the capture system Olympus SC30.

RESULTS

Floral traits in Manihot species and distribution range

We found eight species restricted to the Cerrado and eight more widely distributed or present in other biomes. Floral bracts had diversity in its morphology, being discreet when setaceous and more conspicuous when ovate, though many were deciduous (Table 1, Figure 1). The deciduousness appeared more in the narrow endemics of Cerrado (six out of eight) and less in the wider distributed species (two out of eight). Flowers, in general, were campanulate and all male flowers were campanulate to tubular-campanulate with united tepals (gamotepalous). The female flowers had free

Table 1 Manihot sampled species. Taxa with restricted distribution are indicated as narrow Floral Staining length by Species Distribution Bracts Floral shape (♀ x ♂) (♀ x Voucher neutral ♂, red mm) campanulate (p) x 13.0 x H. Farinasso M. esculenta cultivated setaceous (d) A, M, P campanulate (g) 5.0 A12, UB Amazon, campanulate (p)x 7.0 x H. Farinasso M. tripartita ovate A, M, P Cerrado, campanulate (g) 9.0 A8-9, UB M. campanulate (p) x H. Farinasso Caatinga setaceous 12.0 A, M pseudoglaziovii campanulate (g) A6, UB campanulate (p) x H. Farinasso M. glaziovii Caatinga setaceous 15.0 A, M, P campanulate (g) A5, UB campanulate (p) x 8.0 x M. Mendoza M. fruticulosa Cerrado setaceous A, M, C campanulate (g) 8.0 et al. 4801 H. Farinasso lanceolate campanulate (p) x short- 7.5 x M. gracilis Cerrado A, M, P A7, A10-11, campanulate (g) 8.0 UB ovate- campanulate (p) x 10.8 x H. Farinasso M. violacea Cerrado A, M, P lanceolate campanulate (g) 13.2 A3-4, UB M. Mendoza lanceolate campanulate (p) x tubular- 7.0 x M. congesta Cerrado A, M, C et al. 5294, (d) campanulate (g) 9.5 CEN M. Mendoza Cerrado, setaceous tubular- campanulate (g) x 9.0 x M. debilis A, M, P et al. 4396, narrow (d) tubular- campanulate (g) 12.0 CEN M. Mendoza Cerrado, linear- campanulate (g) x 7.0 x M. minima A, M, P et al. 4405, narrow lanceolate campanulate (g) 7.5 CEN Cerrado, campanulate (p) x 6.8 x H. Farinasso M. oligantha setaceous A, M narrow campanulate (g) 8.6 A1-2, UB linear- M. Mendoza Cerrado, campanulate (p) x 6.0 x M. pinatiloba lanceolate A, M, P et al. 4341, narrow campanulate (g) 9.0 (ned) CEN M. Mendoza M. Cerrado, linear- campanulate (g) x 7.5 x A, M, C et al. 5020, porphyrantha narrow lanceolate (d) campanulate (g) 14.5 CEN ovate- M. Mendoza Cerrado, campanulate (g) x tubular- 8.5 x M. purpurea lanceolate A, M, P et al. 4386, narrow campanulate (g) 10.0 (ned) CEN ovate- M. Mendoza Cerrado, short-campanulate (g) x 8.0 x A, M, M. robusta lanceolate et al. 4354, narrow short-campanulate (g) 10.5 C, P (ned) CEN; UB M. Mendoza Cerrado, subelliptic tubular- campanulate (g) x 9.5 x A, M, M. sagittata et al. 5101, narrow (d) tubular- campanulate (g) 12.5 C, P CEN Abbreviations: bracts deciduousness (d, deciduous; ned, not early deciduous), female (♀) and male (♂) flowers, tepal union (g, gamotepal; p, polytepal), tepal region stained with neutral red (A, apex; M, margins; C, central tepal; P, patchy pattern).

tepals (polytepalous), but six species narrowly distributed in Cerrado featured female flowers with united tepals, similar to male flowers (e.g. M. porphyrantha). We notice that the campanulate shape was variable among species, from short-campanulate, subglobose to tubulose (Fig. 1e, f, h). The flowers measured between 6-15 mm, characterized as small size and male flowers tended to be relatively larger than female flowers (except in M. esculenta). Flowers had green, white, yellow, purple coloration, and sometimes combined colors (result showed for tepals in Fig. 2). However, individuals can differ in flower color (green or yellow, even purple to green) and purplish tinge in the same species. Likewise, bracts can vary among individuals of a species (Fig. 1a, i).

All Manihot species had osmophores in their tepals, with secretory epidermis identified by neutral red staining (Fig. 3). The positive staining with neutral red evidenced two main osmophore patterns on the ventral tepals: 1) diffused secretory epidermis on the tepal surface; 2) papillae on tepal distal margin and apex. The osmophore had rough areas formed by papillae, shining papillae- like trichomes and also trichomes.

The osmophore localization among species was very conservative within flower perianth: evident in the margin and apex of two tepals, discreet and more restricted to the apex in two others, and half of the structure is visible on the last tepal (Fig. 4). Furthermore, certain species only had discreet osmophores (M. gracilis, M. tripartita, Fig. 3a, m), even though they followed the mentioned pattern. These species and also Manihot glaziovii from Caatinga had specific colors within the pattern, e.g. white, green and half white half green (Fig. 2m). The surfaces most enlarged were from osmophores of Cerrado species (Fig. 2b, f-l). Trichomes were a common feature and did not appear only in M. pinatiloba. In general, trichomes were not constant in the same tepal in size

Figure 1 Inflorescences and flowers in Manihot species to illustrated the floral diversity of the genus, all the species occur in Cerrado. a Conspicuous bracts of M. tripartita. b Showy bracts of M. violacea. c Setaceous bracts of M. veadeirensis, with Apis mellifera visiting the male flower. d Discreet bracts (arrow) of M. oligantha. e Wide openned male flowers of M. gracilis. f M. nana campanulate male flower g Tubular-campanulate male flowers of M. incisa. h Tubular male flower of M. tenella. i Female flowers of M. tripartita, note that the bracts do not appear as in (a). j Wide open female flower of M. robusta. k Campanulate female flower of M. incisa with pink stigma. l Intense purple and campanulate female flower of M. porphyrantha.

Figure 2 Tepals (ventral side) in Manihot species. Note the modified epidermis on tepals margins. a M. tripartita. b M. sagittata. c M. debilis. d M. fruticulosa. e M. robusta. f M. purpurea. g M. oligantha. h M. minima. i M. violacea. j M. pinatiloba. k M. congesta. l M. porphyrantha. m M. gracilis. n M. esculenta. o M. glaziovii. p M. pseudoglaziovii. Scale bars: (a-c, e-l, n-o) 0.5 mm, (d, m) 0.2 mm, (p) 1 mm.

(small to long) or branching (one to two). Trichomes frequently demonstrated positive reaction with neutral red (Fig. 3b-c, f, h-i, k-l, n).

Floral nectaries (FNs) were very constant in their location and general morphology among species (not shown). We found secretion on the bottom of both male and female flowers in all studied species. The nectariferous epidermis appeared compact and smooth under the stereoscopic microscope.

Figure 3 Osmophore regions in ventral side of tepals in Manihot species (staining with neutral red). a M. tripartita, details of tepals margins and central region. b M. sagittata, reaction in tepal apex, and trichomes. c M. debilis, detail of patchy epidermis and stained trichome. d M. fruticulosa, positive reaction for apex and central tepal. e M. robusta, in detail central tepal with patchy epidermis. f M. purpurea, note the patchy epidermis (weak reaction) in central of tepal and in detail stained tepals margin. g M. oligantha. h M. minima, in detail the reaction of tepals margin next to a stained trichome. i M. violacea, note in detail the tepals margin positive reaction. j M. pinatiloba. k M. congesta, papillae and trichome-like positive reaction. l M. porphyrantha. m M. gracilis, note the strongest reaction on tepal apex and in detail the patchy epidermis. n M. esculenta, in detail tepals margin and trichomes positive reaction. o M. glaziovii, apex showing the strongest reaction. p M. pseudoglaziovii, note the reaction for margins. Scale bars: (a-j, l-p) 0.5 mm, (details h-i, o) 0.2 mm, (details a-c, e-f, k-n) 0.1 mm.

Osmophores structure

We confirmed the osmophore occurrence based on the papillose epidermis and ruggae with a great number of secretory vesicles, epidermal cells with lipophilic content and terpenoid

Figure 4 Osmophore pattern on the perianth in Manihot species (c, neutral red; d-k, NADI reagent). a, d-g M. oligantha. b, h-k M. violacea. c M. fruticulosa. a Note the round yellowish margins of tepals. b Proeminent tepals margin also yellowish colored. c Pattern on perianth. d Positive reaction for terpenes, revealed by NADI reagent (blue coloration) in tepals margins. e Tepal (hpo) reacting for terpenes in margins, more proeminent in the left side whereas appears the papillae. f Detail of papillae stained by NADI. g Tepal apex (io) that demonstrates restricted terpenes content. h Tepal showing intense reaction with NADI in margins and also in trichomes (detail). i Tepal (io) with the positive reaction for terpenes on margins and central tepal. j Strong reaction for terpenes in papillae on the margins of tepal. k Margins of tepal (io) showing the smooth surface and diffuse terpene content. Abbreviations: (io) inconspicuous osmophore, (hpo) half prominent osmophore, (po) prominent osmophore. Scale bars: (c) 1 mm, (d, e, j) 0.2 mm, (f, g, detail of i, k) 0.1 mm, (h, i) 0.5 mm.

accumulation under the cuticle. The three species were different in the general osmophore structure but have a similar gland location and in the papillae and trichomes components (Fig. 5). The osmophores of Manihot show floral position in the apical part of tepals and in the tepals margins. In

SEM, the osmophores did not vary between sexes of the species. Cerrado species had epidermis projected clefts on the ventral side of the tepals, formed by the papillae (Fig. 5i-j, o-p). M. violacea

Figure 5 Ventral side of tepal in SEM of the Manihot species showing osmophores in apices and margin. a-h M. esculenta. i-n M. violacea. o-t M. oligantha. a General aspect of tepal showing many trichomes. b Margin of tepal with the projected rounded epidermis (papillae) and trichomes. c Tepal’s margin with small projections. d Tepal’s papillae with smooth striation in the cuticle. e, f Detail of the projected epidermis of tepals margins with wrinkled cuticle. g Unusual epidermal projection in the central tepal. h Trichome with peeling. i Tepal showing papillae in margins and trichomes nearby it. j Papillae with a cavity and trichomes. k Papillae with a smooth wrinkled surface. l Detail of tepals margin with wrinkled cuticle. m Projected epidemis that formed the papillae showing slightly striated surface and cuticle peelings. n Trichome with cuticle blisters. o Overall tepal morphology, with rugae and small trichomes within margins. p Tepal margin showing epidermal projections with small clefts and ruptured trichomes (setae). q Detail of epidermal projections, displaying different shapes and a bifurcated structure like-papillae. r Detail of epidermis margin with smooth and projected surface and ruptured trichome. s Cuticle bearing blisters and peeling. t Detail of ruptured trichome containing secretion with circular marks. Scale bars: (a, i, o) 100 µm, (b, c, k) 20 µm, (d, f, g, l, p, q, r) 10 µm, (e, s) 5 µm, (h) 1 µm, (j) 50 µm, (m) 3 µm, (n, t) 2 µm.

had cuticular blisters and slightly rugose surface (Fig. 5k, m) while M. oligantha appeared with a smooth cuticle that showed spherical marks as well as small projections of vesicles secretion (Fig.

5q-s). M. esculenta osmophore appeared different, with a compressed aspect, cuticle striation and scattered epidermal projection (Fig. 5c-f). The trichomes differed in location and density among

29 species: M. esculenta had dense trichomes especially in tepal margin but not in the apex; M. violacea had mainly trichomes near the tepal margin, while M. oligantha had them mainly in the tepal margin

(Fig. 5a, h-i, n-o, r). Many trichomes in this last species appeared ruptured (Fig. 5p, r, t).

The osmophores inner structure in the three Manihot species was formed by secretory cells of the epidermis, cells containing dense vesicles in cytoplasm and trichomes (Fig. 6). The secretory epidermal cells of the osmophores have large amounts of vesicles and dense nucleus (Fig. 6b, d-e, g- h, n) but their structure lack precise organization which appeared from one-layered to multi- multilayered (Fig. 6d, f, o). The osmophore had xylem and phloem vascularization and laticifers appeared at the inner part of the tepal structure. The secretory epidermis was predominantly papillose, though in M. esculenta it tended to palisade in rugae, in osmophore apex (Fig. 6c).

Since idioblasts can be treated as a specialized cell containing a distinctive form, size and chemical composition as described in De Luca et al. (2000) and Castro and Demarco (2008), many cells in Manihot floral glands bear those features and are considered here as idioblasts. Additionally, the idioblasts are a common feature in Euphorbiaceae, occurring in several plants tissues (Metcalfe and Chalk 1950), including in Manihot leaves and fruits (Oliveira and Oliveira 2009, da Cunha Neto et al. 2014). The anatomy revealed a great number of idioblasts in Manihot tepals and the secretory cells of osmophores epidermis often surrounded idioblasts, especially in M. oligantha and M. violacea in the papillae (Fig. 6f, j, l, m, o). Besides that, M. violacea had trichomes-like structures formed by idioblasts or secretory epidermal cells (Fig. 6j, k). The trichomes were mainly uniseriate but could also be composed of two cells (Fig. 6j, p).

Floral nectary structure

The structure of FN was conservative for the three Manihot species (Fig. 7). The morphology of FNs was discoid, the male flower had FN with 10 lobes, whereas female flower had an entirely round nectary, but commonly presented reentrances or lobes in nectary, occasionally with staminate

Figure 6 Anatomy of the osmophore in the tepal margins of Manihot species (stains: a-i toluidine blue 1%, j-p safranin- toluidine blue). a-e M. esculenta. f-j M. violacea. k-o M. oligantha. a Tepal apex with osmophore forming projections in the ventral side. b Osmophore papillae, note the cells with many vesicles. c Osmophore epidermis, composed of secretoty cells and idioblasts. d epidermal projection constituted mainly by idioblasts. e Trichomes (setae) in epidermal projection. f Osmophore apex showing papillae with longer extend ventrally. g Secretory cells of epidermis and idioblasts, note that epidermis can be projected by idioblasts. h Osmophore epidermis, with a large amount of vesicles and an idioblast appear in degradation centrally. i Detail of trichome (setae) in osmophore epidermis. j Secretory cells of osmophores forming a projection similar to trichomes (setae). k Idioblasts with dense content also forming a structure similar to trichome (setae). l, m, n Papillae with longer extend ventrally and large idioblasts. o, p Epidermal projections with idioblasts and secretory cells, note the different trichomes (setae). Scale bars: (a, b, d, e, f, i) 50µm, (c, g, h, j, k, l, m, n, o, p) 30µm.

vestiges (Fig. 7a-b). In SEM, FNs presented compact epidermis with cells of variable shape, often covered by a substance, maybe derivate from the cuticle (Fig. 7e-h). Sparse stomata appeared in mid-down position, near wrinkled areas (Fig. 7c-d, i).

Figure 7 Floral nectaries in Manihot species – morphology (a), micromorphology (SEM, b-k) and anatomy (l-t). a, e, l, o-p M. esculenta. b-d, g-k, m, q-r M. violacea. f, n, s-t M. oligantha. a Nectary of male flower, note the shining nectar that accumulated on the bottom (asterisk). b Lower view of nectary with wrinkling areas and a staminoide. c, d Detail of the previous photo, showing modified stomatas (setae). e, f, g Epidermis of nectary, showing shrink cells. h Epidermis with stigmatic secretion (triangle) from the female flower. i modified stomata with pore occlusion in M. oligantha. j, k detail of not functional stomatas. l, m, n Longitudinal section of male flowers with vascularized nectary except for the M. oligantha. o, p, q Nectary of female flowers in longitudinal section, note the laticifers spread along nectariferous parenchyma. r, s, t Detail from previous sections showing the uniseriate epidermis formed by idioblasts and nectariferous epidermal cells. Scale bars: (a) 1 mm, (b) 100 µm, (c, d, h, i, j, k, o, p, q) 10 µm, (e, f) 30 µm, (g) 20 µm, (l, m, n) 300 µm, (r, s, t) 25 µm.

The internal structure was formed by a nectariferous epidermis single-layered composed of densely stained idioblasts (Fig. 7l-t). These idioblast frequency changes in species, being more

32 recurrent in M. violacea, frequent in M. oligantha and less frequent in M. esculenta. The epidermal cells varied in shape from cubical to rectangular and had a relatively thick cell wall. Nectariferous parenchyma was multilayered and specialized, composed of small secretory cells with a conspicuous nuclei and dense granular cytoplasm, characteristic of floral nectaries (Fahn 1988). FN was vascularized mainly by phloem but also xylem, and vessels had an origin in the receptacle (Fig. 7l- m). We observed 10 traces that ascended to nectary, as earlier described in floral anatomy of M. esculenta (Venkata-Rao and Ramalakshmi 1968). Male flowers of M. oligantha were supplied by those vessels, although it was the only floral type that we did not find vascularization directly in nectariferous parenchyma (Fig. 7m). Laticifers were spread along all FNs and the majority was branched, not articulate.

Floral glands histochemistry

The main chemical compounds found in floral glands did not differ in Manihot species

(Table 2), and between female and male flowers. Terpenes were abundant for both floral glands and all structures had blue coloration indicating essential oils: osmophores, its trichomes, FNs, all had vesicles, whereas idioblasts could present dense homogenous stain (Fig. 8a-c, e, t). We found similar results for NADI (terpenoid reagent) and neutral red. Manihot esculenta had a very strong reaction with NADI (not shown), while M. violacea was moderate and M. oligantha was less accentuated.

The reaction for lipids was more subtle, and occurred in epidermal cells of osmophores and its trichomes (Fig. 8d, f-i) and less often in idioblasts of the glands (Fig. 8i, k). Secretory idioblasts also were positive for phenolic compounds in both glands (Fig. 8e, j, p, q, t).

FNs were positive for starch and carbohydrates that appeared abundant in the parenchyma cells, and only Manihot esculenta presented starch in the nectariferous epidermis, though occasionally (Fig. 8k-l). Depending on the stage of secretion in FNs, the quantity of starch varied. In the beginning, it was abundant, with grains all over the cytoplasm of nectariferous parenchyma cells,

Table 2 Histochemical reactions in the tissues of the floral glands in three Manihot species Floral gland and Compound Secretory structures and species histochemical test M. esculenta M. violacea M. oligantha Osmophore Ep Tr Id Ep Tr Id Ep Tr Id Lugol Starch ------Sudan Black Lipids + - + + + + + - + Sudan IV Lipids + - - + - - + - - Ferric chloride III Phenolic compounds - - + - - + - - + NADI Terpenoids + + + + + + + + +

Floral nectary Ep Pa Id Lat Ep Pa Id Lat Ep Pa Id Lat Lugol Starch + + - - - + - - - + - - PAS Carbohydrates + + + + + + + + + + + + Ruthenium red Acidic mucilages + - - - + - - - + - - - Sudan Black Lipids - - + - - - + - - - + - Ferric chloride III Phenolic compounds - - + - - - + - - - + - Wagner's reagent Alkaloids + + + + + + + + + + + + Comassie R250 Proteins + + + + + + + + + + + + NADI Terpenoids + + + + + + + + + + + +

Notes: (+ positive; - negative). Abbreviations: Ep, epidermis; Tr, trichomes; Id, idioblasts; Pa, parenchyma; Lat, laticifers.

and on more advanced stages there was much less starch. Acid mucilage was detected in cells walls

of the nectariferous epidermis (Fig. 8o). In all FNs tissues, we detected alkaloids and proteins, with

an abundance of alkaloids (Fig. 8r, s).

Figure 8 Histochemical reactions of the floral glands of three Manihot species for osmophores (a-j) and FN (k-t). a, i-k, m, r M. oligantha. b-c, f, l, p, s M. esculenta. d-e, g-h, n-o, q, t M. violacea. a Terpenes (blue coloration) in tepal apex, double-arrow pointing secreted terpene that accumulates below the cuticle. b Margin of tepal bearing papillate epidermis with secreted terpenes. c Trichome showing terpene content. d Papillate epidermis with lipophilic content (red) pointed by double-arrow. e Papillae of the epidermis (double-arrow) and idioblast (triangle) positive for terpene. f Secretory epidermis containing lipids (dark) (double-arrow). g, h, i lipophilic content (dark) in uniseriate trichomes and also epidermis (double-arrow). j Idioblasts (triangle) with phenolic compounds (dark-brown). k Nectariferous parenchyma with starch (dark) near the epidermis. l Nectariferous epidermis positive for starch (losangle). m, n Nectariferous parenchyma, idioblasts (triangle) and laticifers (arrow) with positive reaction for carbohydrates (pink). o The cuticle of the nectariferous epidermis with relatively thick cuticle containing acid mucilage (pink). p Idioblasts (triangle) of nectariferous epidermis reacted positively for phenolic contents (dark-brown). q Discreet reaction of idioblast (triangle) from the nectariferous epidermis for lipid (dark). r Laticifer (arrow), idioblast (triangle) and general reaction for protein detection (blue). s Positive reaction for alkaloids (reddish-brown) in nectariferous tissue. t Nectary containing terpenes (blue) in epidermis and parenchyma. Scale bars: (a, d) 25 µm, (b) 100 µm, (c, e, f, g, h, i, j, l, m, n, o, p, q, r, s, t) 30 µm, (k) 200 µm.

DISCUSSION

Osmophores in Manihot species

Osmophores can be quite common among flowering plants, and are a conspicuous feature in the flowers of all studied Manihot. Though, our new finding for the genus is the second register of this gland for Euphorbiaceae. The study of Gagliardi et al. (2018) registered osmophores in the apex of petals in Joannesia. Here, the osmophores occur in ventral tepals, especially in apex, which is common as described by Vogel (1990); besides, the tepals margins also featured the gland. Although neutral red can fail to identify osmophore in certain cases (Kearns and Inouye 1994, Gonçalves-

Souza et al. 2017), the scent-producing tissues, rich in lipophilic volatile oils, can efficiently react with neutral red, as observed here. The terpenes appeared as the main compound in Manihot osmophores. Floral volatiles is diverse (Jürgens et al. 2008), but terpenes can comprise most of its diversity, being a very common constituent (Dudareva and Pichersky 2000, Knudsen et al. 2006,

Borghi et al. 2017).

The osmophores positioning on Manihot tepals are very conservative for the species and forms a pattern on perianth: margin conspicuous always next to the discreet one. Corolla and/or calyx have mainly two roles: attraction and defense (Delph 1996, Galen 1999, Endress 2010). Likely due to the reduction of Manihot floral whorls, its perianth could do both functions. The osmophore as a chemical barrier for pathogens was observed in a citrus flower (Marques et al. 2015). These researchers also found that floral buds had compact papillae that could represent a structural barrier for antagonists, preventing infection by a pathogen fungus. Here, the papillae on tepals margins in contact with the other inconspicuous side of tepals margins could also function as a structural barrier.

In addition, cells related to defense are in those papillae, such as idioblasts that could protect the floral structure against pathogens and florivory; this protection role is recognized for idioblasts, and is reinforced by other structures found is osmophores such as laticifers, trichomes, and also for the detected substances (e.g phenolic compounds and terpenes) (Mura et al 2005, Hagel et al. 2008,

Demarco and Carmello-Guerreiro 2011, Mithöfer and Boland 2012, Pichersky and Raguso 2016). At anthesis time, the idioblasts have its content mobilized in osmophores, similarly with what was found in Leguminosae (Marinho et al. 2018). However, as demonstrated by these authors, the idioblasts mobilization was not involved in scent production and otherwise could act in plant protection. In short, the consistent patterns of osmophores on Manihot perianth, the identified compounds, and structures, all suggest a pronounced defense function by this floral gland.

The internal structure of osmophore is variable despite the constancy in its perianth position and secretory elements. The variation includes that in the same region the secretion process can be different for each cell: some cells appear with a high amount of vesicles in the cytoplasm, others with vesicles merging with cell walls and other with no vesicles at all. Even the stain differs in a sample, to papillae from another. We might attribute the different secretory stages with a continuum emission of volatiles. The liberation of volatiles by osmophores was described to occur through epidermis cuticle that should be facilitated by cuticle striation (Stern et al. 1987, Vogel 1990, Marinho et al.

2014), but other ways to expose the secretion were reported as stomata (de Melo et al 2010), trichomes diffusion (Wiemer et al. 2009), cuticular pores (Pridgeon and Stern 1983) and cuticular blisters rupture (Sazima et al. 1993). In addition, volatilization by passive diffusion is not possible and the emission process appears complex and challenging to prove (Widhalm et al. 2015, Borghi et al. 2017). Volatile emission is related to increases in the epidermal surface (as papillae, rugae) or to thermogenesis processes (Raguso and Gottsberger 2017, Vogel 1962). In our samples, M. esculenta was the only one with cuticle striation. The epidermis presented cuticle expansion detected solely in

M. violacea, likely due to secretion diffusion, while M. oligantha presented cuticle blisters. Despite the differences, all of them have enlarged areas in tepals margins that should increase scent emission.

Secretory trichomes appear to be very common in tepals of Manihot species. Trichomes can emit floral volatiles (Wiemer 2009), even though their content was showed to not contribute to floral scent and rather act in flower defense (Effmert et al. 2005). In our samples, they seemed to work

37 together with epidermis for terpene emission, but they also work independently. It is known that the production of conical cells in petals of Antirrhinum (Plantaginaceae) has its pathway in common with the trichomes (Glover et al. 1998), therefore the epidermis and trichomes could be connected in its development. The trichome-like papillae could function as the epidermis and emit terpenoids, whereas the uniseriate trichome (more differentiated) might function for deterrence. Although, M. violacea species has uniseriate trichome that appears non-glandular fullith lipophilic content and terpenes and M. esculenta trichomes are also rich in terpenes. A connection between these trichomes and the secretory epidermis is likely, but the function of these former structures is still unclear.

Floral nectaries in Manihot species

The floral nectaries of the studied Manihot species are highly conservative in its structure, secretory components, and chemical compounds. The epidermis has few modified stomata located whereby nectar accumulates, suggesting that stomata secrete nectar. Despite that, wrinkled areas occur in floral nectary during anthesis, at the same time the epidermis appear narrower and can be degenerated. Due to the scattered stomata and epidermal modification after secretion, probably nectariferous epidermis also releases nectar. It is known that nectar secretion by stomata can be a non regulated process (Razem and Davis 1999, Wist and Davis 2006), even though several authors attributed for modified stomata the nectar release (Davies et al. 2005, Paiva and Machado 2008,

Stpiczynska et al. 2014), including other Crotonoideae (Euphorbiaceae) as Joannesia and Croton

(Gagliardi et al. 2016, Feio et al. 2016). Similarly with those in Crotonoideae, in Manihot species, floral nectary has single-layered nectary epidermis, but Croton did not present differentiated epidermal cells strongly stained and Joannesia contained chlorophyll in nectariferous parenchyma, which differentiates Manihot nectary.

Related to its origin, nectar can be secreted in small quantities at a time (chlorophyll) or rapidly when starch is stored in the parenchyma, often its storage being source to nectar syntheses

(Pacini et al. 2003, Razem and Davis 1999, Durkee et al. 1981). The Manihot species floral nectaries largely contain starch grains that disappear depending on the stage of secretion. In field observation, nectar release starts hours before flower opening, despite that we only observed floral visitors attracted to the floral bud near to its opening. The process indicates that the starch is stored in the parenchyma, working as the precursor of nectar, then, in pre-anthetic flowers occur the nectar release and during floral anthesis, the nectar metabolism slows down. Flower of M. esculenta was reported to last one day (Halsey et al. 2008). Similarly, the nectar strategy of Manihot species appears to be ephemeral, consisting of offering this reward already when the flower open for the first time, and afterward little or no nectar should be produced.

Many substances related to plant defense are found in nectary tissue of Manihot species, such as phenolics, alkaloids, and terpenes (review in Mithöfer and Boland 2012). Cassava stems, roots and leaves contain these mentioned compounds that can act in the defense system (Blagbrough et al.

2010, Thiyagarajan and Suriyavathana 2010, Zeng et al. 2015). In plants damaged by herbivory, the alkaloid levels are positively correlated between nectar and leaves (Adler et al. 2006). Besides the protection function, alkaloids in nectar might affect pollination, including repel illegitimate visitors and shape the dynamics of pollinators visits, promoting outcrossing (Kessler and Baldwin 2007,

Kessler et al. 2008, Irwin and Adler 2008). It is likely that nectar in Manihot contains alkaloids that could enhance reproductive fitness and flower defense.

Floral diversification in Manihot species and evolutionary implications

As discussed previously, osmophore seems to have a dual role of protection and attraction due to its structure and compounds. Plant defense system often evolved first, with a floral attractant or reward evolving secondarily, as suggested for monoterpenes, resins, fragrance and inflorescence color (Pellmyrl and Thien 1986, Armbruster et al. 2009, Hanley et al. 2009, Schiestl 2010). Despite, evolutionary switches in function occurred for triterpene resin, which started as a toxin and deterrent

39 compound, switching to an attractive floral reward and then turning back for the protection role

(Armbruster et al. 1997). In addition, modifying the concentration of terpenes may shift its functions

(Theis and Lerdau 2003) or terpenoids compounds can even have a double function of attraction and defense (Knudsen et al. 2006). These findings all indicate a lability of terpenes. Here, we implied that the osmophore first evolved for defense due to antagonist pressure and, afterward, could function in pollinator advertisement by the emission of essential oils (terpenes).

From osmophore consistent pattern, we also propose that its disposal has an economic value and would be a sort of cheaper investment. Considering that the genus is predominantly monoecious and depends on animal pollination (Alves 2002), reproductive features could be related to certain constraints of this sexual system. It is known that it is more costly to produce unisexual flowers

(Jong et al. 2008), this might explain why the genus flowers are generally small sized and have perianth in a single whorl (least costly). Taxa bearing unisexual flowers are predicted to have generalist pollination (Bawa 1980, Charlesworth 1993, Jong et al. 2008). In fact, the morphology of both male and female flowers in Manihot is among the highest generalization level and enables most visiting animals (Olesen et al. 2007). The sexual dimorphism is predominantly in the genus, and the tendency is the perianth length in male flowers being bigger than female flowers. It is known that female flowers reduced size is relative to the hermaphroditic ancestor of related groups on Lycium

(Miller and Venable 2003). Allocation of resources to male and female functions is related to sexual selection (Willson 1979) and a smaller female might save resources to allocate in fruit development

(Huang et al. 2006).

For bracts features an economy seems to exist as well, especially in the narrow endemic species that most have deciduous or very inconspicuous bracts (e.g. M. oligantha). Lavergne et al.

(2004) explain in their work on the Mediterranean region that narrow endemics invest less in reproduction than its widespread congeners due to the need for local persistence on the occupation of rocky soils. Here, the narrow studied species often occur in this soil condition (see Mendoza et al.

2015, 2016) and deciduous bracts might save resources. It is known that bracts have both attraction and defense function (Sun et al. 2008, Armbruster et al. 2009, Endress 2010). For the narrow endemic species is more likely that these structures protect the buds than function in attractiveness. If showy, the bract of Manihot could act in pollinator attraction, even though few species here presents attractive bracts (M. purpurea, M. tripartita, M. violacea). As a result, the perianth should often be the major verticil with attraction function in species.

In Manihot genus the male flowers are campanulate or tubular-campanulate shape, whereas the female flowers have free tepals and are wide-opening (see Rogers and Appan 1973). Differently, from eight species restricted distribution in the Cerrado, six had the female flowers with connected tepals, comparing with the wider-distributed species, none have this feature. The female flowers of restricted endemic species also show the smallest flowers. It is known that changes in flower size can correspond to changes in the members’ size of a functional group that pollinates the flowers (Galen

1999, Fenster et al. 2004). We suggest that the Cerrado species evolved differential floral traits revealed by the morphology of the female flowers, which might be adapted to their local pollinator fauna, thus indicating more specialization of species on this biome.

Furthermore, specialized morphology of osmophores occurs in Cerrado taxa, wherein it can be very similar to pollen (Fig. 4b), and actually, bees are attracted to the structure and scrape it

(personal observations in M. violacea). The reported pollinators for Manihot (Nassar and Carvalho

1990) are mainly generalist social bees, in which the learning of flower signals is potentialized by flower odors (Kunze and Gumbert 2001, Leonard 2011). Even generalist pollinators might exert divergent selection in flower shape, in a population under strong selective regimes (Gómez 2009).

Manihot species have a noteworthy geographic clustering in drier climates, but considering the efforts, its phylogeny is still not well resolved, probably due to its plasticity, ancestral polymorphism, and ongoing speciation (Chacón et al. 2008, Duputié et al. 2011, Cervantes-Alcayde et al. 2015).

Despite that, our analyzed species with showy osmophores occurs in subclades of Cerrado (Mendoza

2018, unpublished phylogeny). In contrast, the species that are more widely distributed and the ones from Caatinga biome bear inconspicuous osmophores. Therefore, indicating that cerrado species diversified in its glands.

Although certain floral traits appear constant such as perianth in a single whorl, nectary and osmophores position, the individual variation of flower color, glands morphology and trichomes are characteristic of Manihot species, which we could imply phenotypic plasticity of the genus. By plasticity, the conservative flowers may respond to the environment and monoecy constraints. The floral glands in Manihot and its secondary compounds could play an important adaptive role since perianth structure kept constant in whorls and tepal number. However, a full analysis of the secretory process and floral volatiles still need to be conducted. We emphasized that the attraction or deterrent effect are closely related and may overlap functions.

REFERENCES

Adler, L. S., Wink, M., Distl, M., & Lentz, A. J. (2006). Leaf herbivory and nutrients increase nectar alkaloids. Ecology Letters, 9, 960–967. https://doi.org/10.1111/j.1461-0248.2006.00944.x

Alves, A. A. C. (2002). Cassava Botany and Physiology. Chapter 5 in: Cassava: Biology, Production and Utilization, CABI, Wallingford, United Kingdom., 67–89.

An, F., Chen, T., Mouafi, D., Stéphanie, A., Li, K., & Li, Q. X. (2016). Domestication syndrome is investigated by proteomic analysis between cultivated cassava (Manihot esculenta crantz) and its wild relatives. PLoS ONE, 1–34. https://doi.org/10.1371/journal.pone.0152154

Armbruster W. S. (1984). The role of resin in angiosperm pollination: Ecological and chemical considerations. American Journal of Botany, 71:1149–1160.

Armbruster, W. S. (1985). Patterns of character divergence and the evolution of reproductive ecotypes of Dalechampia scandens (Euphorbiaceae). Evolution, 39(4), 733–752.

Armbruster, W. S. (1997). Exaptations link evolution of plant-herbivore and plant-pollinator interactions: a phylogenetic inquiry. Ecology, 78(September), 1661–1672. https://doi.org/10.1890/0012-9658(1997)078[1661:ELEOPH]2.0.CO;2

Armbruster, W. S., Lee, J., & Baldwin, B. G. (2009). Macroevolutionary patterns of defense and pollination in Dalechampia vines: Adaptation, exaptation, and evolutionary novelty. Proceedings of the National Academy of Sciences, 106(43), 18085–18090. https://doi.org/10.1073/pnas.0907051106

Bawa, S. H. (1980). Evolution of dioecy in flowering plants. Annual Review of Ecology and Systematics, 11, 15–39.

Bergougnoux, V., Caissard, J. C., Jullien, F., Magnard, J. L., Scalliet, G., Cock, J. M. Baudino, S. (2007). Both the adaxial and abaxial epidermal layers of the rose petal emit volatile scent compounds. Planta, 226(4), 853–866. https://doi.org/10.1007/s00425-007-0531-1

Bernardello, G. (2007). A systematic survey of floral nectaries. In Nectaries and Nectar. Springer (pp. 19–128).

Blagbrough, I. S., Bayoumi, S. A. L., Rowan, M. G., & Beeching, J. R. (2010). Cassava: An appraisal of its phytochemistry and its biotechnological prospects. Phytochemistry, 71(17–18), 1940–1951. https://doi.org/10.1016/j.phytochem.2010.09.001

Borghi, M., Fernie, A. R., Schiestl, F. P., & Bouwmeester, H. J. (2017). The Sexual Advantage of Looking, Smelling, and Tasting Good: The Metabolic Network that Produces Signals for Pollinators. Trends in Plant Science, 22(4), 338–350. https://doi.org/10.1016/j.tplants.2016.12.009

Byrne, D. (1984). Breeding Cassava. Chapter 3 in: Plant Breeding Reviews, Vol. 2.

Cacho, N. I., Berry, P. E., Olson, M. E., Steinmann, V. W., & Baum, D. A. (2010). Are spurred cyathia a key innovation? Molecular systematics and trait evolution in the slipper spurges (Pedilanthus clade: Euphorbia, Euphorbiaceae). American Journal of Botany, 97(3), 493–510. https://doi.org/10.3732/ajb.0900090

Carmo Júnior, J. E., Sodré, R. C., & Silva, M. J. (2013). Manihot (Euphorbiaceae s.s.) no Parque Estadual da Serra Dourada, Goiás, Brasil. Rodriguésia, 64(4), 727–746.

Castro, M. M., & Demarco, D. (2008). Phenolic compounds produced by secretory structures in plants: a brief review. Natural Product Communications, 3(8), 1273–1284.

Cervantes-Alcayde, M.-A., Olson, M. E., Olsen, K. M., & Eguiarte, L. E. (2015). Apparent similarity, underlying homoplasy: Morphology and molecular phylogeny of the North American clade of Manihot. American Journal of Botany, 102(4), 520–532. https://doi.org/10.3732/ajb.1500063

Charlesworth, D. (1993). Why are unisexual flowers associated with wind pollination and unspecialized pollinators? The American Naturalist, 141(3), 481–490.

Da Cunha Neto, I. L., Martins, F. M., Caiafa, A. N., & Martins, M. L. L. (2014). Leaf anatomy as subsidy to the taxonomy of wild Manihot species in Quinquelobae section (Euphorbiaceae). Brazilian Journal of Botany, 37, 481–494. https://doi.org/10.1007/s40415-014-0057-2

David. R., Carde J. P. (1964). Coloration différentielle dês inclusions lipidique et terpeniques dês pseudophylles du Pin maritime au moyen du reactif Nadi. Comptes Rendus Hebdomadaires dês Séances de l’ Academie dês Sciences Paris. Série D 258, 1338–1340.

Davies, K. L., Stpiczyńska, M., & Gregg, A. (2005). Nectar-secreting floral stomata in Maxillaria anceps Ames & C. Schweinf. (Orchidaceae). Annals of Botany, 96(2), 217–227. https://doi.org/10.1093/aob/mci182 de Melo, M. C., Borba, E. L., & Paiva, E. A. S. (2010). Morphological and histological characterization of the osmophores and nectaries of four species of Acianthera (Orchidaceae: Pleurothallidinae). Plant Systematics and Evolution, 286, 141–151. https://doi.org/10.1007/s00606-010-0294-1

Delph, L. F., Galloway, L. F., & Stanton, M. L. (1996). Sexual dimorphism in flower size. The American Naturalist, 148(2), 299–320. https://doi.org/10.1086/285926

Demarco, D. (2017). Floral glands in asclepiads: structure, diversity and evolution. Acta Botanica Brasilica, 31(3), 477–502. https://doi.org/10.1590/0102-33062016abb0432

Demarco, D., & Carmello-Guerreiro, S. M. (2011). Pericarp ontogeny and histochemistry of the exotesta and pseudocaruncle of Euphorbia milli (Euphorbiaceae). Rodriguésia, 62(3), 477– 489.

De-Paula, O. C., das Sajo, M. G., Prenner, G., Cordeiro, I., & Rudall, P. J. (2011). Morphology, development and homologies of the perianth and floral nectaries in Croton and Astraea (Euphorbiaceae-Malpighiales). Plant Systematics and Evolution, 292(1–2), 1–14. https://doi.org/10.1007/s00606-010-0388-9

Domínguez, C. A., Dirzo, R., & Bullock, S. H. (1989). On the Function of Floral Nectar in Croton suberosus (Euphorbiaceae). Oikos, 56(1), 109–114. https://doi.org/10.2307/3566093

Dudareva, N., & Pichersky, E. (2000). Biochemical and Molecular Genetic Aspects of Floral Scents. Plant Physiology, 122(3), 627–634. https://doi.org/10.1104/pp.122.3.627

Durkee, L. T., Gaal, D. J., Reisner, W. H., Durkee, T., Gaal, D. J., & Reisner, W. H. (1981). The floral and extra-floral nectaries of Passiflora. I. The floral nectary. American Journal of Botany, 68(4), 453–462.

Effmert, U., Große, J., Röse, U. S. R., Ehrig, F., Kägi, R., & Piechulla, B. (2005). Volatile composition, emission pattern, and localization of floral scent emission in Mirabilis jalapa (Nyctaginaceae). American Journal of Botany, 92(1), 2–12. https://doi.org/10.3732/ajb.92.1.2

Endress, P. K. (2010). Flower Structure and Trends of Evolution in Eudicots and Their Major Subclades. Annals of the Missouri Botanical Garden, 97(4), 541–583. https://doi.org/10.3417/2009139

Erbar, C. (2014). Nectar secretion and nectaries in basal angiosperms, magnoliids and non-core eudicots and a comparison with core eudicots. Plant Diversity and Evolution, 131, 63–143. https://doi.org/10.1127/1869-6155/2014/0131-0075

Fahn, A. (1988). Secretory tissues in vascular plants. New phytologist, 108:229-257.

Fahn, A. (2000). Structure and function of secretory cells. Advances in Botanical Research, 31, 37– 75.

Feio, A. C., Riina, R., & Meira, R. M. S. A. (2016). Secretory Structures in Leaves and Flowers of Two Dragon’s Blood Croton (Euphorbiaceae): New Evidence and Interpretations. International Journal of Plant Sciences, 177(6), 511–522. https://doi.org/10.1086/685705

Fenster, C. B., Armbruster, W. S., Wilson, P., Dudash, M. R., & Thomson, J. D. (2004). Pollination Syndromes and Floral Specialization. Annual Review of Ecology, Evolution, and Systematics, 35(1), 375–403. https://doi.org/10.1146/annurev.ecolsys.34.011802.132347

Fisher, D. B. (1968). Protein staining of ribboned epon sections for light microscopy. Histochemie, 16(1), 92–96. https://doi.org/10.1007/BF00306214

Freitas, L., Bernadello, G., Galetto, L., & Paoli, A. A. S. (2001). Nectaries and reproductive biology of Croton sarcopetalus (Euphorbiaceae). Botanical Journal of the Linnean Society, 136, 267– 277. https://doi.org/10.1006/bojl.2000.0437

Furr, M., Mahlberg, P. G. (1981) Histochemical analyses of laticifers and glandular trichomes in Cannabis sativa. Journal of Natural Products 44, 153-159.

Gagliardi, K. B., Cordeiro, I., & Demarco, D. (2016). Protection and attraction: Bracts and secretory structures in reduced inflorescences of Malpighiales. Flora: Morphology, Distribution, Functional Ecology of Plants, 220, 52–62. https://doi.org/10.1016/j.flora.2016.02.003

Galen, C. (1999). Why Do Flowers Vary? BioScience, 49(8), 631–640. https://doi.org/10.2307/1313439

Glover, B. J., Perez-Rodriguez, M., & Martin, C. (1998). Development of several epidermal cell types can be specified by the same MYB-related plant transcription factor. Development (Cambridge, England), 125(17), 3497–3508. https://doi.org/10.1017/CBO9781107415324.004

Gómez, J. M., Perfectti, F., Bosch, J., & Camacho, J. P. M. (2009). A geographic selection mosaic in a generalized plant — pollinator — herbivore system. Ecological Monographs, 79(2), 245–263.

Gonçalves-Souza, P., Schlindwein, C., Dötterl, S., & Paiva, E. A. S. (2017). Unveiling the osmophores of Philodendron adamantinum (Araceae) as a means to understanding interactions with pollinators. Annals of Botany, 119(4), 533–543. https://doi.org/10.1093/aob/mcw236

Gregory, M., & Baas, P. (1989). A survey of mucilage cells in vegetative organs of the dicotyledons. Israel Journal of Botany, 38(2–3), 125–174. https://doi.org/10.1080/0021213X.1989.10677119

Hagel, J. M., Yeung, E. C., & Facchini, P. J. (2008). Got milk? The secret life of laticifers. Trends in Plant Science, 13(12), 631–639. https://doi.org/10.1016/j.tplants.2008.09.005

Hanley, M. E., Lamont, B. B., & Armbruster, W. S. (2009). Pollination and plant defence traits co- vary in Western Australian Hakeas. New Phytologist, 182, 251–260. https://doi.org/10.1111/j.1469-8137.2008.02709.x

Hershey, C. (2010). A Global Conservation Strategy for Cassava (Manihot esculenta) and Wild Manihot Species. CIAT, A Summary of Stakeholder Deliberations and Recommendations.

Huang, M., Sanchez-Moreiras, A. M., Abel, C., Sohrabi, R., Lee, S., Gershenzon, J., & Tholl, D. (2012). The major volatile organic compound emitted from Arabidopsis thaliana flowers, the sesquiterpene (E)-β-caryophyllene, is a defense against a bacterial pathogen. New Phytologist, 193(4), 997–1008. https://doi.org/10.1111/j.1469-8137.2011.04001.x

Huang, S., Tang, L., Sun, J., & Lu, Y. (2006). Pollinator response to female and male floral display in a monoecious species and its implications for the evolution of floral dimorphism. New Phytologist, 171, 417–424.

Irwin, R. E., & Adler, L. S. (2008). Nectar secondary compounds affect self-pollen transfer: implications for female and male reproduction. Ecology, 89(8), 2207–2217.

Irwin, R. E., Adler, L. S., & Brody, A. K. (2004). The dual role of floral traits: Pollinator attraction and plant defense. Ecology, 85(6), 1503–1511. https://doi.org/10.1890/03-0390

Johansen, D. A. (1940). Plant microtechnique. McGraw-Hill Book Co, New York.

Johnson, M. T. J., Campbell, S. A., & Barrett, S. C. H. (2015). Evolutionary Interactions Between Plant Reproduction and Defense Against Herbivores. Annual Review of Ecology, Evolution, and Systematics, 46, 191–213. https://doi.org/10.1146/annurev-ecolsys-112414-054215

Jong, T. J., Shmida, A., & Thuijsman, F. (2008). Sex allocation in plants and the evolution of monoecy. Evolutionary Ecology Research, 10(8), 1087–1109. https://doi.org/10.1038/149354c0

Junker, R. R., & Blüthgen, N. (2010). Floral scents repel facultative flower visitors, but attract obligate ones. Annals of Botany, 105(5), 777–782. https://doi.org/10.1093/aob/mcq045

Junker, R. R., Gershenzon, J., & Unsicker, S. B. (2011). Floral Odor Bouquet Loses its Ant Repellent Properties After Inhibition of Terpene Biosynthesis. Journal of Chemical Ecology, 37(12), 1323–1331. https://doi.org/10.1007/s10886-011-0043-0

Jürgens, A., Dötterl, S., Liede-Schumann, S., & Meve, U. (2008). Chemical diversity of floral volatiles in Asclepiadoideae-Asclepiadeae (Apocynaceae). Biochemical Systematics and Ecology, 36(11), 842–852. https://doi.org/10.1016/j.bse.2008.10.005

Karnovsky M. J. 1965. A formaldehyde-glutaraldehyde fixative of high osmolarity for use in electron microscopy. Journal of Cell Biology 27, 137–138.

Kearns C. A., Inouye D. W. (1994). Techniques for pollination biologists. University Press of Colorado, Niwot, Colorado.

Kessler, A., & Halitschke, R. (2009). Testing the potential for conflicting selection on floral chemical traits by pollinators and herbivores: predictions and case study. Functional Ecology, 23(5), 901–912. https://doi.org/10.1111/j.1365-2435.2009.01639.x

Kessler, D., & Baldwin, I. T. (2007). Making sense of nectar scents: the effects of nectar secondary metabolites on floral visitors of Nicotiana attenuata. The Plant Journal, 49, 840–854. https://doi.org/10.1111/j.1365-313X.2006.02995.x

Kessler, D., Gase, K., & Baldwin, I. T. (2008). Field experiments with transformed plants reveal the sense of floral scents. Science, 321(5893), 1200–1202. https://doi.org/10.1126/science.1160072

Knudsen, J. T., Eriksson, R., Gershenzon, J., & Stahl, B. (2006). Diversity and Distribution of Floral Scent. The Botanical Review, 72(1), 1–120. https://doi.org/10.1663/0006- 8101(2006)72[1:dadofs]2.0.co;2

Knudsen, J. T., Tollsten, L., & Ervik, F. (2001). Flower scent and pollination in selected neotropical palms. Plant Biology, 3(6), 642–653. https://doi.org/10.1055/s-2001-19366

Kowalkowska, A. K., Kozieradzka-Kiszkurno, M., & Turzyński, S. (2014). Morphological, histological and ultrastructural features of osmophores and nectary of Bulbophyllum wendlandianum (Kraenzl.) Dammer (B. section Cirrhopetalum Lindl., Bulbophyllinae Schltr., Orchidaceae). Plant Systematics and Evolution, 301(2), 609–622. https://doi.org/10.1007/s00606-014-1100-2

Kruckeberg, A. R., & Rabinowitz, D. (1985). Biological aspects of endemism in higher plants. Annual Review of Ecology and Systematics, 16, 447–479.

Kunze, J., & Gumbert, A. (2001). The combined effect of color and odor on flower choice behavior of bumble bees in flower mimicry systems. Behavioral Ecology, 12(4), 447–456. https://doi.org/10.1093/beheco/12.4.447

Lavergne, S., Thompson, J. D., Garnier, E., & Debussche, M. (2004). The biology and ecology of narrow endemic and widespread plants: a comparative study of trait variation in 20 congeneric pairs. Oikos, 107(3), 505–518.

Leonard, A. S., Dornhaus, A., & Papaj, D. R. (2011). Flowers help bees cope with uncertainty: signal detection and the function of floral complexity. Journal of Experimental Biology, 214(1), 113– 121. https://doi.org/10.1242/jeb.047407

Lillie R. D. (1965). Histopathologic technic and practical histochemistry, 3rd Ed. McGraw-Hill Book Co, New York.

Luca, V. De, & Pierre, B. S. (2000). The cell and developmental biology of alkaloid biosynthesis. Trends in Plant Science, 5(4), 168–173.

Lucas-Barbosa, D. (2016). Integrating Studies on Plant – Pollinator and Plant – Herbivore Interactions. Trends in Plant Science, 21(2), 125–133. https://doi.org/10.1016/j.tplants.2015.10.013

Machado, S. R., Paleari, L. M., Paiva, É. A. S., & Rodrigues, T. M. (2015). Colleters on the Inflorescence Axis of Croton glandulosus (Euphorbiaceae): Structural and Functional Characterization. International Journal of Plant Sciences, 176(1), 86–93. https://doi.org/10.1086/678469

Majetic, C. J., Raguso, R. A., & Ashman, T. L. (2009). The sweet smell of success: Floral scent affects pollinator attraction and seed fitness in Hesperis matronalis. Functional Ecology, 23(3), 480–487. https://doi.org/10.1111/j.1365-2435.2008.01517.x

Marinho, C. R., Souza, C. D., Barros, T. C., & Teixeira, S. P. (2014). Scent glands in legume flowers. Plant Biology, 16(1), 215–226. https://doi.org/10.1111/plb.12000

Marinho, C., Martucci, M. E., Gobbo-Neto, L., & Teixeira, S. P. (2018). Chemical composition and secretion biology of the floral bouquet in legume trees. Botanical Journal of the Linnean Society, (March), 1–21. https://doi.org/10.1093/botlinnean/boy002/4947795

Marques, J. P. R., Amorim, L., Silva-Junior, G., Spósito, M. B., & Appezzato-da Gloria, B. (2015). Structural and biochemical characteristics of citrus flowers associated with defence against a fungal pathogen. AoB PLANTS, 7(1), 1–10. https://doi.org/10.1093/aobpla/plu090

McManus, J. F. A. (1948). Histological and histochemical uses of periodic acid. Stain Technology, 23(3), 99–108. https://doi.org/10.3109/10520294809106232

Médail, F., & Baumel, A. (2018). Using phylogeography to define conservation priorities: the case of narrow endemic plants in the mediterranean basin hotspot. Biological Conservation, 224, 258– 266. https://doi.org/10.1016/j.biocon.2018.05.028

Mendoza F., J. M., & Martins, M. L. L. (2018). Distinctive New Species of Manihot (Euphorbiaceae) from Brazil. Novon, 26, 59–65. https://doi.org/10.3417/D1700007

Mendoza F., J. M., Simon, M. F., & Cavalcanti, T. B. (2015). Three new endemic species of Manihot (Euphorbiaceae) from the Chapada dos Veadeiros, Brazil. Arnaldoa, 22(2), 297–312.

Mendoza F., J. M., Simon, M. F., Arquelão, T. K. M., & Cavalcanti, T. B. (2018). Novas espécies de Manihot (Euphorbiaceae) do Brasil Central. Rodriguésia, 69(2), 915–932. https://doi.org/10.1590/2175-7860201869244

Mendoza F., J. M., Simon, M. F., Reis, T. S., Fidelis, A., & Cavalcanti, B. (2016). New endemic species of Manihot (Euphorbiaceae) from Serra do Tombador in Goiás, Central Brazil. Phytotaxa, 273(3), 147–157. https://doi.org/10.11646/phytotaxa.273.3.1

Metcalfe, C. R., & Chalk, L. (1950). Anatomy of the dicotyledonss: Leaves, stem and wood in relation to taxonomy with notes on economic uses, vol. I. Clarendon Press, Oxford, UK.

Miller, J. S., & Venable, D. L. (2003). Floral Morphometrics and the Evolution of Sexual Dimorphism in Lycium (Solanaceae). Evolution, 57(1), 74–86.

Mithöfer, A., & Boland, W. (2012). Plant Defense Against Herbivores: Chemical Aspects. Annual Review of Plant Biology, 63(1), 431–450. https://doi.org/10.1146/annurev-arplant-042110- 103854

Mura, A., Medda, R., Longu, S., Floris, G., Rinaldi, A. C., & Padiglia, A. (2005). A Ca2+/calmodulin-binding peroxidase from Euphorbia latex: Novel aspects of calcium-hydrogen peroxide cross-talk in the regulation of plant defenses. Biochemistry, 44(43), 14120–14130. https://doi.org/10.1021/bi0513251

Narbona, E., & Dirzo, R. (2010). A reassessment of the function of floral nectar in Croton suberosus (Euphorbiaceae): a reward for plant defenders and pollinators. American Journal of Botany, 97(4), 672–679. https://doi.org/10.3732/ajb.0900259

Nassar, N. M. A., Mendoza, F. J. M., & Sano, N. (2012). Some interesting cassava cultivars: 9 – ICB 300. Geneconserve, 11(43), 22–25.

Neiland, M. R. M., & Wilcock, C. C. (1998). Fruit set, nectar reward, and rarity in the Orchidaceae. American Journal of Botany, 85(12), 1657–1671. https://doi.org/10.2307/2446499

Nepi, M., Aderkas, P. Von, Wagner, R., Mugnaini, S., Coulter, A., & Pacini, E. (2009). Nectar and pollination drops: How different are they? Annals of Botany, 104, 205–219. https://doi.org/10.1093/aob/mcp124

Olesen, J. M., Dupont, Y. L., Ehlers, B. K., & Hansen, D. M. (2007). The openness of a flower and its number of flower-visitor species. Taxon, 56(3), 729–736. https://doi.org/10.2307/25065857

Oliveira, J. H. G. de, & Oliveira, D. M. T. (2009). Morfoanatomia e ontogênese do pericarpo de Manihot caerulescens Pohl e M. tripartita Müll. Arg. (Euphorbiaceae). Revista Brasileira de BotâNica, 32(1), 117–129. https://doi.org/10.1590/S0100-84042009000100012

Pacini, E., Nepi, M., & Vesprini, J. L. (2003). Nectar biodiversity: a short review. Plant Systematics and Evolution, 238, 7–21. https://doi.org/10.1007/s00606-002-0277-y

Paiva, E. A. S., & Machado, S. R. (2008). The floral nectary of Hymenaea stigonocarpa (Fabaceae, Caesalpinioideae): Structural aspects during floral development. Annals of Botany, 101(1), 125– 133. https://doi.org/10.1093/aob/mcm268

Paiva, J. G. A., Fank-de-Carvalho, S. M., Magalhães, M. P., & Graciano-Ribeiro, D. (2006). Verniz vitral incolor 500®: uma alternativa de meio de montagem economicamente viável. Acta Botanica Brasilica, 20(2), 257–264.

Parachnowitsch, A. L., Raguso, R. A., & Kessler, A. (2012). Phenotypic selection to increase floral scent emission, but not flower size or colour in bee-pollinated Penstemon digitalis. New Phytologist, 195(3), 667–675. https://doi.org/10.1111/j.1469-8137.2012.04188.x

Pearse, A. G. E. (1985). Histochemistry: theoretical and applied. 4th ed. C. Livingstone, Edinburgh.

Pellmyr, O., & Thien, L. B. (1986). Insect Reproduction and Floral Fragrances: Keys to the Evolution of the Angiosperms? Taxon, 35(1), 76–85.

Perera, P. I. P., Quintero, M., Dedicova, B., Kularatne, J. D. J. S., & Ceballos, H. (2013). Comparative morphology, biology and histology of reproductive development in three lines of Manihot esculenta Crantz (Euphorbiaceae : Crotonoideae). AoB PLANTS, 1–18. https://doi.org/10.1093/aobpla/pls046

Pichersky, E., & Gershenzon, J. (2002). The formation and function of plant volatiles: Perfumes for pollinator attraction and defense. Current Opinion in Plant Biology, 5(3), 237–243. https://doi.org/10.1016/S1369-5266(02)00251-0

Pichersky, E., & Raguso, R. A. (2016). Why do plants produce so many terpenoid compounds? New Phytologist, 220(3), 692–702. https://doi.org/10.1111/nph.14178

Piechulla, B., & Pott, M. B. (2003). Plant scents - Mediators of inter-and intraorganismic communication. Planta, 217(5), 687–689. https://doi.org/10.1007/s00425-003-1047-y

Płachno, B. J., Stpiczyńska, M., Davies, K. L., Świątek, P., & de Miranda, V. F. O. (2017). Floral ultrastructure of two Brazilian aquatic-epiphytic bladderworts: Utricularia cornigera Studnička and U. nelumbifolia Gardner (Lentibulariaceae). Protoplasma, 254(1), 353–366. https://doi.org/10.1007/s00709-016-0956-0

Płachno, B. J., Świątek, P., Stpiczyńska, M., & Miranda, V. F. O. (2018). Flower palate ultrastructure of the carnivorous plant Genlisea hispidula Stapf with remarks on the structure and function of the palate in the subgenus Genlisea (Lentibulariaceae). Protoplasma, 1–8. https://doi.org/10.1007/s00709-018-1220-6

Poinar, G., & Poinar, G. (2018). The antiquity of floral secretory tissues that provide today’s fragrances. Historical Biology, 1–6. https://doi.org/10.1080/08912963.2018.1502288

Possobom, C. C. F., Guimarães, E., & Machado, S. R. (2015). Structure and secretion mechanisms of floral glands in Diplopterys pubipetala (Malpighiaceae), a neotropical species. Flora: Morphology, Distribution, Functional Ecology of Plants, 211, 26–39. https://doi.org/10.1016/j.flora.2015.01.002

Pridgeon, A. M., & Stern, W. L. (1983). Ultrastructure of Osmophores in Restrepia (Orchidaceae). American Journal of Botany, 70(8), 1233–1243.

Raguso, R. A. (2008). Wake Up and Smell the Roses: The Ecology and Evolution of Floral Scent. Annual Review of Ecology, Evolution, and Systematics, 39(1), 549–569. https://doi.org/10.1146/annurev.ecolsys.38.091206.095601

Raguso, R. A., & Gottsberger, G. (2017). An ode to osmophores: Stefan Vogel’s seminal contributions to the study of scent. Flora, 232, 150–152. https://doi.org/10.1016/j.flora.2017.02.014

Razem, F. a, & Davis, A. R. (1999). Anatomical and ultrastructural changes of the floral nectary of Pisum sativum L. during flower development. Protoplasma, 206(1–3), 57–72.

Rogers, D. J., & Appan, S. G. (1973). Manihot Manihotoides (Euphorbiaceae). Flora Neotropica, 13, 1–272.

Sazima, M., Vogel, S., Cocucci, A., & Hausner, G. (1993). The perfume flowers of Cyphomandra (Solanaceae): Pollination by euglossine bees, bellows mechanism, osmophores, and volatiles. Plant Systematics and Evolution, 187(1–4), 51–88. https://doi.org/10.1007/BF00994091

Schiestl, F. P. (2010). The evolution of floral scent and insect chemical communication. Ecology Letters, 13(5), 643–656. https://doi.org/10.1111/j.1461-0248.2010.01451.x

Seigler, D. S. (1994). Phytochemistry and Systematics of the Euphorbiaceae. Annals of the Missouri Botanical Garden, 81(2), 380–401.

Simon, M. F., Grether, R., Queiroz, L. P. De, Skema, C., Pennington, R. T., & Hughes, C. E. (2009). Recent assembly of the Cerrado, a neotropical plant diversity hotspot, by in situ evolution of adaptations to fire. Proceedings of the National Academy of Sciences of the United States of America, 106(48), 20359–20364.

Simpson, B. B., & Neff, J. L. (1981). Floral rewards: alternatives to pollen and nectar. Annals of the Missouri Botanical Garden, 68(2), 301–322.

Sletvold, N., Moritz, K. K., & Ågren, J. (2015). Additive effects of pollinators and herbivores result in both conflicting and reinforcing selection on floral traits. Ecological Society of America.

Solereder, H. 1908. Systematic anatomy of the dicotyledons. Oxford: Clarendon Press. 1182 p.

Steinmann, V. W., & Porter, J. M. (2002). Phylogenetic relationships in Euphorbieae (Euphorbiaceae) based on its and ndhf sequence data. Annals of the Missouri Botanical Garden, 89(4), 453–490. https://doi.org/10.2307/3298591

Stern, W. L., Curry, K. J., & Pridgeon, A. M. (1987). Osmophores of Stanhopea (Orchidaeceae). American Journal of Botany, 74, 1323–1331.

Stpiczynska, M., Nepi, M., & Zych, M. (2014). Nectaries and male-biased nectar production in protandrous flowers of a perennial umbellifer Angelica sylvestris L. (Apiaceae). Plant Systematics and Evolution. https://doi.org/10.1007/s00606-014-1152-3

Strauss, S. Y. (1997). Floral Characters Link Herbivores, Pollinators, and Plant Fitness. Ecology, 70(6), 1640–1645. https://doi.org/doi:10.1890/0012-9658(1997)078[1640:FCLHPA]2.0.CO;2

Sun, J.-F., Gong, Y. B., Renner, S. S., & Huang, S. Q. (2008). Multifunctional Bracts in the dove tree Davidia involucrata (Nyssaceae: Cornales): rain protection and pollinator attraction. The American Naturalist, 171(1), 119–124. https://doi.org/10.1086/523953

Theis, N., & Lerdau, M. (2003). The Evolution of Function in Plant Secondary Metabolites. International Journal of Plant Sciences, 164(May 2003).

Thiyagarajan, M., & Suriyavathana, M. (2010). Phytochemical and Antimicrobial Screening of Manihot Esculanta Crantz Varieties Mulluvadi I, CO3. International Journal of Biotechnology and Biochemistry, 6(6), 859–864.

Thomas, V. (1991). Structural, functional and phylogenetic aspects of the colleter. Annals of Botany, 68, 287-305.

Thornburg, R. W., Carter, C., Powell, A., Mittler, R., Rizhsky, L., & Horner, H. T. (2003). A major function of the tobacco floral nectary is defense against microbial attack. Plant Systematics and Evolution, 238(1–4), 211–218. https://doi.org/10.1007/s00606-003-0282-9

Tong, Z., Wang, X., Wu, L., & Huang, S. (2018). Nectar supplementation changes pollinator behaviour and pollination mode in Pedicularis dichotoma: implications for evolutionary transitions. Annals of Botany, 1–8. https://doi.org/10.1093/aob/mcy102

Veiga, T., Guitián, P., Guitián, J., Guitián, J., & Sobral, M. (2015). Are pollinators and seed predators selective agents on flower color in Gentiana lutea ? Evolutionary Ecology, 29, 451– 464. https://doi.org/10.1007/s10682-014-9751-6

Venkata-Rao, C., & Ramalakshmi, T. (1968). Floral anatomy of the Euphorbiaceae–I. Some non- cyathium taxa. Journal of the Indian Botanical Society, 47, 278–300.

Vitarelli, N. C., Riina, R., Caruzo, M. B. R., Cordeiro, I., Fuertes-Aguilar, J., & Meira, R. M. S. A. (2015). Foliar secretory structures in Crotoneae (Euphorbiaceae): Diversity, anatomy, and evolutionary significance. American Journal of Botany, 102(6), 833–847. https://doi.org/10.3732/ajb.1500017

Vogel, S. (1962). Duftdrüsen im Dienste der Bestaübung. Über Bau und Funktion der Osmophoren. Abhandlungen der Mathematisch-Naturwissenschaftlichen, Akademie der Wissenschaften, Mainz, 10, 1–165.

Vogel, S. (1990). The role of scent glands in pollination. On the structure and function of osmophores. Amerind Publishing, New Delhi.

Vogel, S., & Hadacek, F. (2004). Contributions to the functional anatomy and biology of Nelumbo nucifera (Nelumbonaceae) III. An ecological reappraisal of floral organs. Plant Systematics and Evolution, 249(3–4), 173–189. https://doi.org/10.1007/s00606-004-0203-6

Webster, G. L. (1994). Classification of the Euphorbiaceae. Annals of the Missouri Botanical Garden, 81(1), 3–32.

Widhalm, J. R., Jaini, R., Morgan, J. A., & Dudareva, N. (2015). Rethinking how volatiles are released from plant cells. Trends in Plant Science, 20(9), 545–550. https://doi.org/10.1016/j.tplants.2015.06.009

Wiemer, A. P., Moré, M., Benitez-Vieyra, S., Cocucci, A. A., Raguso, R. A., & Sérsic, A. N. (2009). A simple floral fragrance and unusual osmophore structure in Cyclopogon elatus (Orchidaceae). Plant Biology, 11, 506–514. https://doi.org/10.1111/j.1438-8677.2008.00140.x

Willson, M. F. (1979). Sexual selection in plants. The American Naturalist, 113(6), 777–790.

Wiśniewska, N., Kowalkowska, A. K., Kozieradzka-Kiszkurno, M., Krawczyńska, A. T., & Bohdanowicz, J. (2018). Floral features of two species of Bulbophyllum section Lepidorhiza Schltr.: B. levanae Ames and B. nymphopolitanum Kraenzl. (Bulbophyllinae Schltr., Orchidaceae). Protoplasma, 255(2), 485–499. https://doi.org/10.1007/s00709-017-1156-2

Wist, T. J., & Davis, A. R. (2006). Floral nectar production and nectary anatomy and ultrastructure of Echinacea purpurea (Asteraceae). Annals of Botany, 97(2), 177–193. https://doi.org/10.1093/aob/mcj027

Wright, G. A., & Schiestl, F. P. (2009). The evolution of floral scent: The influence of olfactory learning by insect pollinators on the honest signalling of floral rewards. Functional Ecology, 23(5), 841–851. https://doi.org/10.1111/j.1365-2435.2009.01627.x

Zeng, Y. B., Li, S. S., Mei, W. L., Dong, W. H., Li, K. M., & Dai, H. F. (2015). Two new terpenoids from the stems of Manihot esculenta. Journal of Asian Natural Products Research, 17(3), 280– 284. https://doi.org/10.1080/10286020.2014.958996

CHAPTER 2

Ecological generalized but not so much: pollinator filtering in monoecious

Manihot (Euphorbiaceae)

Hannah C. Farinasso¹, Hélder Consolaro², Antonio J. C. Aguiar³

¹Programa de Pós-Graduação em Botânica, Universidade de Brasília, DF, Brasil.

²Instituto de Biotecnologia, Universidade Federal de Goiás, Regional Catalão, GO, Brasil.

³Departamento de Zoologia, Universidade de Brasília, DF, Brasil.

Manuscript to be submitted to the Journal Arthropod-Plant Interactions

ABSTRACT

Starting from the available fauna of floral visitors to them performing as pollinators, several floral filters might act. Here, we questioned if the unisexual flowers can function as filter, wherein visitors go only for one floral type. Besides, we reported for the first time the pollination and breeding system of wild Manihot in its natural environment. We studied two Manihot species, and for each one recorded the visitors and pollinators in two flowering events. We evaluated differences among floral sexes such as floral anthesis and display, floral morphometrics, nectar volume and concentration; also, we conducted hand-pollination treatments. Overall, the Manihot species were generalist and attracted several groups of insect visitors, but less were recorded pollinating. Manihot violacea showed ecological specialization with bees, whereas M. oligantha had two main functional groups (i.e. ants and flies). The floral display sex ratio was male-biased and the functional groups that are low abundant tend to not visit female flowers. Despite, there was a tendency of the female flower being more attractive in nectar traits and it was not dicriminated by pollinators. The Manihot species depended on animal pollination to produce seeds and the relative abundance of pollinators contributes to its importance and possible selection of floral phenotypes. Each species have distinctive traits in nectar, floral anthesis, and display that might restrict pollinators. Moreover, the female flowers exert filtering in occasional visitors by its lower number in these monoecy systems.

The findings reinforce that unspecialized flowers can restrict pollinators in different ways than floral morphology.

Key words actinomorphic flower, cassava wild relatives, cerrado, endemic species, female-biased nectar, pollination syndromes, sexual dimorphism

INTRODUCTION

In pollination systems, specialization and generalization is a currently debated topic (Brosi

2016). For instance, the concepts of pollination syndrome are being criticized and its predictions have been put to test in nature, yet, evidence seems to support the syndromes overarching patterns and predictions (Wilson et al. 2004, Martén-Rodríguez et al. 2010, Padyšáková et al. 2013, Rosas-

Guerrero et al. 2014, Abrahamczyk et al. 2017, Wilson et al. 2017). Simultaneously, generalist interactions plant-pollinators are suggested to be a broad pattern (Waser et al. 1996, Lázaro et al.

2008, Ollerton et al. 2009, Reverté et al. 2016). Despite the controversies, it is well accepted that the diversity on plants of these systems ranges from extreme specialized to supergeneralist, in a complex continuum (Waser et al. 1996, Johnson and Steiner 2000, Joly et al. 2018). Moreover, a different pollination system might be attributed, depending on the scale (time and space), functional category

(functional or taxonomic groups) or even methodology (Fenster et al. 2004, Petanidou et al. 2008, Li and Huang 2009, Ollerton et al. 2015).

It seems to exist a dichotomy between specialist—generalist pollination, e.g.: actinomorphic flowers are associated with generalist pollination, whereas zygomorphic flowers are thought to be specialized (Neal et al. 1998, Fenster et al. 2004, Sargent et al. 2004, Gong and Huang 2009). Many authors (cited above) argue that flowers with restricted access could be more specialized than wide open flowers. Nevertheless, the openness of a flower has not shown to be correlated to generalization levels (Herrera 1988, Olesen et al. 2007). In addition, other studies suggest that actinomorphic flowers might be more specialized than zygomorphic depending on the community context (Lázaro et al. 2008) and that symmetry may not be related to degrees of specialization (Ashworth et al.

2015).

Floral morphology that restricts visitor access is not the only indicator of specialization in flowers. In flowers with open rewards, specialization can occur by other floral traits, such as nectar composition that filter certain floral visitors (Johnson et al. 2006, Herrera et al. 2010), selective

56 olfactory cues (Shuttleworth and Johnson 2010, Burger et al. 2017), nectary marks (Wang et al.

2018), and anthesis phenology (Martén-Rodríguez et al. 2009). In plants that feature separated male and female phases, filtering of potential pollinators might occur, wherein the pollinator needs to visit both phases to deposit pollen in the stigma (Armbruster 2017). Consequently, in a monoecy system, the pollinator should visit both male and female flowers. Furthermore, it is known that female flowers must be discriminated by pollinators (Charlesworth 1993, Ashman 2000, Totland 2004,

Etten and Chang 2014), thus, these floral types might act as a floral filter.

According to Rosas-Guerrero et al. (2014) findings, floral syndromes were better predicted in tropical plants and in taxa that rely on animal pollination for reproduction (monoecious, dioecious, or self-incompatible systems). Manihot is a neotropical genus, predominantly monoecious and outcrossing system (Rogers and Appan 1973, Da Silva et al. 2003, Lai et al. 2014). Therefore, in

Manihot syndromes or specialization is expected. The Manihot flowers are actinomorphic with exposed nectary that appears to offer no barrier for visitors, and the only reported pollinators are hymenopterans, among eusocial wasps and bees (Faegri and Pijl 1979, Kawano 1980, Rogers and

Appan 1973, Nassar and Carvalho 1990); however, most pollination studies of the genus were in cultivars. Even though there is an urgency to conserve crop wild relatives due to its possible contributions to crop improvement and helping global food security (Vincent et al. 2013), e.g. the wild species M. oligantha have great potential to increase protein content through hybrids with cassava (Nassar et al. 2012), the pollination system of wild Manihot species remains unexplored.

Here, we studied two monoecious species of Manihot with the following objectives: (1) evaluate the specialization-generalization of the pollination system concerning the array of floral visitors, pollinators and functional groups; (2) verify the dependence on animal pollination and the existence of pollen limitation by exploring reproductive traits of breeding systems; (3) investigate if monoecy could act as a filter for pollinators. For this, we studied floral traits related to pollinator filtering such as differences in floral anthesis and display, rewards and morphology in male and female flowers.

Furthermore, we checked functional groups that visited only one floral type or did not contact reproductive verticils of both male and female flowers, thus resulting in filtering.

MATERIAL AND METHODS

Study species and populations

We studied two species of Manihot endemic to Brazilian Cerrado: M. violacea Pohl and M. oligantha Pax. The main center of diversity and endemism of the genus is in Cerrado, therein

Manihot species often present narrow distribution in this biome (Rogers and Appan 1973, Simon et al. 2018). The primary site that occurs M. oligantha is in the Private Reserve of Natural Heritage

Serra dos Topazios (-16.737399, -47.689693), Cristalina-GO. The vegetation has predominately presence of herbaceous vegetation, subshrubs, and spaced trees. For M. violacea we assessed two localities, in Brasilia-DF, Dom Bosco Ecological Park (-15.796089, -47.805440) and a slope Cerrado area popular named ‘Córrego do Urubu’ (-15.7182440, -47.8518380). These areas have vegetation which mostly consists of Cerrado sensu stricto with shrubs, herbaceous and trees.

The Manihot flowers are conservative with single whorl in pentamerous perianth and have protogyny at the inflorescence level (Rogers and Appan 1973). The two studied species differ in plant size length, inflorescence traits, bracts, and flower color. Manihot violacea is a subshrub ca. 1.0 m tall, with greenish flowers and numerous conspicuous foliaceous bracts that form a compact inflorescence (Figure 1A-C). The second, M. oligantha is nearly acaulescent subshrub ca 10 cm tall, with purplish flowers, inconspicuous setaceous bracts and inflorescences that are elevated above the vegetative part (Fig. 1D-F). Both species present strong floral dimorphism, with polytepalous female flowers (FF) and gamotepalous male flowers (MF).

Floral anthesis and display

Figure 1 The studied Manihot species showing their habitat, inflorescences, and unisexual flowers. A-C, M. violacea, and D-F, M. oligantha.

We followed the floral anthesis for 30 marked buds in pre-anthesis, for each flower sex, in a period of three consecutive days. The presence of floral rewards (pollen, nectar) and floral movements (anthers contraction and retraction, tepals positioning) helped us for flower functioning evaluation. For daily floral display, we counted the total number of male and female flowers

(Yakimowski et al. 2011). For this, we did preliminary phenological observations and established 30 random individuals for samplings, and assessed the population over one year to record a flowering season (May 2017 to April 2018).

Breeding system and dependence on pollinators

For testing the mating system we conducted artificial pollination for four treatments: geitonogamy (pollen of the same plant), xenogamy (pollen from other plant), apomixes as asexual reproduction (isolated female flowers) and open pollination (natural), all in previously bagged

59 female flowers for pollen control (except natural pollination). Geitonogamy was only performed when male and female flowers from the same plant were open on the same day. For the xenogamy, we collected pollen and mixed from random individuals at a minimum distance of 10m. We selected female flowers that opened in the day of the treatments. The experiment period was from December

2017 to January 2018. We collected the treatments that developed fruit for seeds count and for analysis of the seed quality through the weight of the fresh seeds. An index of self-incompatibility

(ISI) was obtained by dividing the mean number seed set per flower from the treatments of geitonogamy by those from the xenogamy: self-compatible species score 1, incompletely compatible species values between 1 and 0, and self-incompatible species score below 0.2 (Zapata and Arroyo

1978).

Floral morphometrics and nectar traits

For floral measurements, we obtained 32 flowers for each floral type, of different individuals.

The collected fresh flowers were photographed using a scale, in three different angles (frontal, lateral and lateral exposing reproductive structures). We used software ImageJ to analyze the images. The measurements were linear, consisted of perianth diameter (opposite distance from tip to tip of tepals), perianth length, tepals junction (floral tube), tepal width, flower bottom width, floral nectary diameter and height, and floral pedicel.

Considering that nectar traits are quite responsive to environmental variation and the genetic heritable variation can be lost under field conditions (Mitchell 2004), we sampled nectar in the two first hour of anthesis to diminish environment interference. Each flower had nectar sampling made once, from different individuals in previously bagged flowers, totalizing 10 flowers per sex. The nectar volume was measured with 5 µL micro capillaries for M. violacea and 1 µL micro capillaries for M. oligantha and the sugar nectar concentration by using a hand refractometer (Atago 0–90%

Brix).

Floral visitors, pollinators and functional groups

We performed observation of floral visitors and the entomological collection alternating each activity for 30 minutes, for two consecutive days during the flowering period. Each species had a total of 60h of sampling: in M. violacea was 40h for Dom Bosco population in November 2017 to

February 2018, and 20h for Urubu population in January 2018; for M. oligantha we divided observation into two flowering periods, 40h from November 2017 to February 2018, and 20h from

December 2018 to January 2019. The observation hour was equal for FF and MF. The data from the visits consisted of morphospecies, visiting time, floral type, presence of pollen from previous visits, contacts with reproductive floral parts, behavior and accessed floral resource. We collected specimens directly on flowers and analyzed under stereoscopic for taxon determination and observation of pollen on the insect body. The pollen of Manihot is easy to be recognized due to the very large size, spherical shape and the exine featuring pila ornaments (see Vieira et al. 2012).

Specimens are deposited in the Entomological Collection of the Universidade de Brasília (DZUB)

(Appendix).

We treated as legitimate visits when floral visitors touched the reproductive verticils and carried pollen; the potential pollinators were the insects that visited legitimately and besides it, visited male and female flowers. Further, we grouped the floral visitors and pollinators into functional groups (FG), defined basically as visitors that interacted similarly with the flowers

(Fenster et al. 2004, Gómez 2009). For FG establishment, we applied criteria of foraging behavior, feeding habits, proboscis length, and body size. We grouped floral visitors into 11 functional groups: ants, bees, flies, grasshoppers, large bees, large beetles, microlepidoptera, short-tongued bees, small bees, small beetles and wasps.

Functional groups importance and specialization of the pollination system

For the estimation of importance in each FG of pollinator, we followed the equation proposed by Armbruster (2014): pollinator importance (PI) = V ×A×S, in which V is the visitation rate per unit time, A is the per-visit probability of contacting the anthers and S is the per-visit probability of contacting the stigmas.

We based on Fenster et al. (2004) for the analysis of the specialization-generalization degree that FG that representing three fourths of floral visits frequency can be considered specialized in this

FG. Although, these authors discussed that if two FG likely exerts similar selection pressure, specialization can be considered. In contrast, if two FG of pollinators are clearly selecting to different traits due to their habitat, e.g. bats and hummingbirds, this plant species is rather a generalist (Martén-Rodríguez et al. 2009).

Statistical analysis

The floral traits related to measurements and nectar (volume and sugar concentration) had a comparison between sexes in each species by one-way ANOVA, through the packages ‘car’ and

‘carData’ (Fox and Weisberg 2011, 2018). Likewise, to test if there were differences among pollination treatments, the seed weight means per fruit followed the mentioned test above. The seed set per fruit were compared between pollination treatments that followed a GLM model with Poisson distribution, by using the package ‘emmeans’ (Lenth 2018). To compare the number of visits to each floral type, we used the chi-square test of goodness-of-fit, with the floral display means ratio as the given probabilities. We created the graphs by using the package ‘ggplots2’ (Wickham 2016). All the statistical analysis was run in R environment v. 3.5.1 (R Core Team 2018).

RESULTS

Anthesis and floral display sex ratio

Floral anthesis was similar between sexes and species, wherein flowers lasted one day. The anthesis was throughout daylight, longer in M. oligantha since the flowers opened between 6-7 am and were senescent by 6 pm, while M. violacea flowers opened between 11-12 am and also were senescent by 6 pm. The bud in pre-anthesis releases nectar that accumulates inside flowers, on the perianth basis, and if this floral bud was positioned against natural or artificial light, it reflected nectar drops. The male bud had already exposed the pollen when the nectar accumulated. The first movements of perianth are rapid, with centrifugal curling of the tepals. However, the returning movements are slow and in male flowers are characterized when the anthers contract inwards. In M. oligantha the perianth started to close midday, with contraction of the anthers at 2 pm. Similar, M. violacea contracted anthers at 3 pm.

The flowering occurred during the rainy season of Cerrado (from November to February) for the studied populations. The two sampled populations had male-biased floral display: in M. violacea the number of males to female flowers per day was always higher, and had sex ratio of 25 male flowers to 5 female flowers on average; Manihot oligantha had a less male-biased floral display, with sex ratio means of 25 male flowers to 18 female flowers. Generally, the number of flowers displayed per individuals was low, and occasionally the sexes overlapped in the same individual

(asymmetric protogyny).

Breeding system and dependence on pollinators

The species showed to be dependent on pollen transference to produce seed, thus there was no seed in apomixes treatments (Table 1). The seed set per fruit was similar between treatments and the cross and open pollination treatments were not statically different, therefore pollen limitation was not detected for the plant populations. The treatment geitonogamy had lesser seed set per fruit but not statically different from the others. Seeds of geitonogamy had less weight than seeds of natural

Table 1 Seed set and seed weight means ± standard deviation across treatments and sample sizes in parentheses Seed set per fruit Seed weight (mg) Index Species Open Cross Self Apomixis Open Cross Self ISI 2.20 ± 2.65 ± 2.15 ± 60.50 ± 64.20 ± 61.70 ± 0 M. violacea 0.77 0.48 0.67 14.38 16.62 14.96 0.81 (n=39) (n=33) (n=38) (n=32) (n=53) (n=69) (n=43) 2.38 ± 2.86 ± 2.11 ± 34.20a ± 31.11ab± 27.50b ± 0 M. oligantha 0.76 0.51 0.78 5.56 3.86 5.97 0.74 (n=46) (n=56) (n=35) (n=27) (n=31) (n=43) (n=19) Only treatments that show significant difference present superscript letters (in bold, at Tukey’s honestly significant difference, t=-2.884, P=0.01).

pollination in M. oligantha, revealing that the seeds quality of this former treatment were the lowest.

Both species were incompletely compatible according to ISI index.

Dimorphism in floral morphometrics and in nectar traits

The flowers were small sized, and the species had bigger MF concerning floral length and

tepal width, and in M. oligantha, MF was also larger in perianth width (Table 2). The campanulate

MF had a short tube, whereas the polytepalous FF many times opened partly, resulting in a shorter

tube that functioned as campanulate-shaped. Consequently, perianth diameter size (tepals tip to tip)

tent to not differ between the sexes, though FF of M. violacea presents a weak tendency to be bigger

in diameter. Floral pedicel of FF was wider in both species. The nectary of the species differed

between sexes: M. oligantha had MF with wider nectary diameter, in contrast, in this species FF had

a taller nectary; in M. violacea there was only a tendency of nectary in FF to be taller (Table 2).

The nectar traits had differences between sexes and were particular to each species (Figure

2). M. oligantha presented a markedly sexual dimorphism for nectar, with higher volume and

concentration in FF. For M. violacea, the nectar volume was alike between MF and FF, but sugar

concentration had a tendency to be higher in FF.

Table 2 Floral measurements and comparison between floral types showing means ± standard deviation M. violacea M. oligantha Morphometric traits Female Male F Female Male F Perianth diameter 17.64 ± 4.47 15.88 ± 3.37 3.22 . 11.23 ± 1.74 11.35 ± 1.78 0.07 Perianth lenght 10.82 ± 1.37 13.12 ± 1.54 39.68*** 6.87 ± 0.73 8.61 ± 1.16 51.7*** Perianth tube 3.70 ± 1.63 8.62 ± 1.56 151.7*** 2.39 ± 0.88 5.52 ± 0.97 96.66*** Perianth width 6.16 ± 0.93 5.90 ± 0.97 1.19 5.30 ± 0.54 6.24 ± 0.78 31.09*** Tepal width 5.04 ± 0.68 5.61 ± 0.86 8.31** 3.69 ± 0.35 4.31 ± 0.42 40.12*** Nectary diameter 4.56 ± 0.48 4.66 ± 0.67 0.52 4.06 ± 0.24 4.55 ± 0.42 30.72*** Nectary height 1.57 ± 0.20 1.47 ± 0.19 3.87 . 1.15 ± 0.08 1.10 ± 0.09 5.40* Pedicel 2.74 ± 0.55 1.37 ± 0.29 156.7*** 1.31 ± 0.16 1.18 ± 0.15 35.01*** F test significance, 0.07>. P>0.05 marginal significance * P>0.01, ** P>0.001, *** P>0.0001 (n=32 for floral type).

Figure 2 Nectar traits of M. violacea in A-B, and M. oligantha, C-D. Different letters are significantly different at P<0.05 by F-test.

Generalization of the pollination system and temporal specialization

Practically all the 11 analyzed FG of insect visitors touched anthers and carried pollen, also

opportunistic insects contact stigma surface, therefore being legitimate visitors (Table 3).

Table 3 Functional groups recorded on Manihot flowers Plant Functional Body Order Families and Floral Behavioural notes Type of species group length genera resources visit M. ants variable Hymenoptera Formicidae nectar long visits; have a legitimate violacea (Camponotus, protective behavior on + Dorymyrmex, female flowers that illegitimate Pseudomyrmex) appears to defend the nectar bees variable Hymenoptera Apidae nectar + nectaring between tepals, legitimate (Ceratina, pollen but also hangs on stigma + Partamona, to nectaring; collects illegitimate Trigona) pollen

flies variable Diptera Muscidae land on stigma legitimate

grasshoppers variable Orthoptera nectar nymphs legitimate

large beetles > 7 mm Coleoptera Curculionidae nectar introduces part of the legitimate head in the flowers large bees > 10 Hymenoptera Apidae pollen entering with ventral part legitimate mm (Bombus) of body in the flower; rapid visit short- variable Hymenoptera Halictidae nectar + pollen collecting; legitimate tongued (Augochlora, pollen explored and gathering + bees Augochloropsis, on stigma illegitimate Paraxystoglossa, Rhinocorynura)

small beetles < 3 mm Coleoptera Chrysomelidae nectar long visits in which stay legitimate on the flower bottom + robbing nectar; copulated illegitimate in flowers

small bees < 3 mm Hymenoptera Apidae (Plebeia) nectar explored the flower illegitimate without carrying pollen microlepido- variable Lepidoptera nectar exploing the nectar legitimate ptera around tepals

wasps variable Hymenoptera Vespidae nectar uses stigma as land legitimate (Polybia) surface and hangs on it + for nectaring illegitimate

Table 3 (Continued) Plant Functional Body Order Families and Floral Behavioural notes Type of species Group length genera resources visit

M. ants variable Hymenoptera Formicidae nectar long visits; legitimate oligantha (Camponotus, constantly leaves the + Dorymyrmex, flowers and returns; illegitimate Ectatomma, agressive with other Pseudomyrmex, visitors Solenopsis) flies variable Diptera Chloropidae nectar + land on stigma and legitimate (Liohippelates), pollen attracted to the + Empididae stigma surface; in illegitimate (Empis), male fowers can Lonchaeidae, consume pollen Muscidae (Musca, Sarcopromusca) grasshoppers variable Orthoptera nectar + nymphs legitimate pollen short- variable Hymenoptera Halictidae nectar + nectaring on tepals, legitimate tongued (Dialictus) pollen but touches stigma + bees sometimes; collects illegitimate pollen small < 3 mm Coleoptera Megalopodidae nectar long visits; consume legitimate beetles pollen

wasps variable Hymenoptera Vespidae nectar uses stigma as land legitimate (Polybia) surface and hangs on it for nectaring

Hymenopterans that formed FG of ants, short-tongued bees and wasps, occurred in both Manihot

species and the flies FG. The two Manihot species had in common three genera of floral visitors: the

social wasp Polybia and the ants Pseudomyrmex and Camponotus (Table 3). Overall, the floral

visitors and pollinators were generalist insects (Figure 3).

We found differences among FG in M. violacea populations and between the two years of

study in M. oligantha (Figure 4). In M. oligantha, FG of ants and flies represented a maximum of

72% and 68% in a flowering, thus the population appeared not specialized in either FG. In M.

violacea, FG of bees represented 74.50% on average of total visits and both studied populations had

Figure 3 Different functional groups of pollinators on Manihot flowers, M. violacea A-D, and M. oligantha E-H. A, small beetle (Chrysomelidae); B, short-tongued bee (Augochloropsis); C, bees (Partamona cupira); D, detail of P. cupira cobiculae with M. violacea pollen; E, ants (Ectatomma brunneum); F, grasshopper (nymph); G, fly (Empis) carrying pollen (arrow) from M. oligantha; H, fly (Liohippelates) detail with M. oligantha pollen (arrow). Scale bars 0.1 mm.

Figure 4 Functional groups of floral visitors for the Manihot species, across floral seasons (years 2017-2019) in the study population. Data are expressed as the percentage of total visits for a given flowering season in each population.

FG of bees as the most representative: 94.20% and 66.26% of total visits. Furthermore, 90% of total

FG of bees was from a single stingless bee species, the Partamona cupira. This bee species could locate the rewarding flower prior to its opening and searched mainly for nectar and less frequently collected pollen (10% of total visits). Partamona cupira was the only bee species that during floral anthesis appeared exclusively carrying M. violacea pollen (Figure 3). The constancy of this bee in

M. violacea was noteworthy in both sampled populations. As a result, an ecological short-term specialization between the social native bee P. cupira and M. violacea can be considered at least on the ‘Urubu’ population.

The pollination system had a high level of generalization, as we found six and five functional groups (FG) of potential pollinators for M. violacea and M. oligantha, respectively (Figure 5). FG of pollinators that appeared the most important were the social bees and wasps for M. violacea and the ants and flies for M. oligantha. Although, FG that we have seen transferring pollen to stigma were:

Figure 5 Functional groups of potential pollinators showing their pollinator importance (PI) to each Manihot species, combining the sampled sites and flowering seasons.

the bees (n=39), wasps (n=9) and short-tongued bees (n=1) in M. violacea, and the flies (n=4) and wasps (n=2) in M. oligantha.

Floral visitors and pollinator filtering

Observations revealed that six FG of floral visitors only visited male flowers (Figure 6). The relative abundance of floral visitors shown to be a filter in M. violacea, in which visitors with less than 0.10 of abundance did not visit female flowers, with the exception of the flies. In M. oligantha, the majority of FG visited male and female flowers, though the small beetles FG, which had the lowest abundance did not visit female flowers among the two years. Despite that, several FG of pollinators, including the most important, showed a higher frequency in female flowers than expected (Figure 7), thus this floral type was not dicriminate.

Figure 6 Relative abundance of functional groups among male and female flowers in M. violacea A, and M. oligantha B. Functional groups that do not visit one floral type are represented in grey with a black line.

We registered Apis mellifera in field excursions, in all study areas, however it did not visit

Manihot flowers. Likewise, stingless bees and solitary bees in the region of M. oligantha were identified, though we observed only a short-tongued bee (Dialictus sp.) visiting the flowers, in later morning (ca. 11:00). Altogether, the findings indicate an attractant filter of the local fauna (sensu

Armbruster 2017).

Figure 7 Total visits of functional groups to male and female flowers and the corresponded chi-squared value, in M. violacea A, and M. oligantha B. Asterisks indicate statistical difference of X2 values among floral types, for each functional group (* P>0.01, ** P>0.001, *** P>0.0001). Only functional groups of pollinators with five or more visits were included.

DISCUSSION

Generalization and specialization in Manihot species

Considering the number of floral visitors and functional groups that potentially pollinates the species, the system tends to be ecological and functionally generalized. Overall, these pollinators are

72 extremely generalist, varying from social bees and wasps to flies, ants, and grasshoppers. The stamens are arranged in a way that all floral visitors touched it and carried pollen, with the exception of very small ants and bees. Even opportunistic insect groups that appeared with very low relative abundance (e.g grasshoppers), touched stigma. Generalization is advantageous when pollinator assemblage is variable, leading to reproduction assurance (Perfectti et al. 2009, Martén-Rodríguez et al. 2010) and even though the species show dependence on animal pollination, they are not pollen limited. Considering the heterogeneity of Cerrado environment (Silva et al. 2006, Dodonov et al.

2014), the generalist systems could be an advantage. The exclusive presence of three ovules in

Manihot likely contributes to the reproduction assurance and pollination by unspecialized and insects. Furthermore, generalized plants bearing a few ovules did not show difference among their visitors effectiveness-quality component (Gómez and Zamora 1999, Zych et al. 2018). In fact, the quantitative component can have a central role in plant reproduction rather than the qualitative component (Vázquez et al. 2005, Sahli and Conner 2006) and the most abundant is often the most important pollinators.

Although the pollination system is generalized, the association of Manihot with social hymenopterans was strong as both species had bees and ants as the most important pollinators.

Therefore, the importance of these social insects for the Manihot can be prominent, likely due to their abundance. Indeed, an ecological specialization of short-term could be established between a native stingless bee and M. violacea in at least one population. The eusocial bees are known to be related to dioecious and monoecious plants, in which they forage constantly at a specific plant for a while (Bawa 1977, 1980). The Partamona bee has high floral fidelity to M. violacea that might be related to its unusual foraging strategy of mass-recruiting (Flaig et al. 2016). The behavior of

Partamona can lead to dominance over resources, mainly by the size of a colony that excludes other species (Hrncir and Maia-Silva 2013), in this case, the nectar resource of both unisexual flowers that should be highly attractive for them.

Ants were the most abundant visitors in M. oligantha. This species has many traits of plants pollinated by ants, covering nine out ten traits that Hickman (1974) predicts: low height, dense populations (see Simon et al. 2018) with intermixed individuals that often grow in places poorly vegetated, small flowers with accessible nectar, between others. The ant pollination is still seen as rare, despite there is crescent evidence of its occurrence and importance (de Vega et al. 2009, de

Vega and Gómez 2014, Ibarra-Isassi and Sendoya 2016, Domingos-Melo et al. 2017; for a different perspective see Rostás et al. 2018). Recently, the ant syndrome was unveiled in a Cerrado species - the studied species possessed similar habitat of M. oligantha, and the ant agents were shown to be the most effective pollinators (Del-Claro et al 2019). In addition to this qualitative trait, pollination by ants has the quantity component as a key factor, in which the system relies on the abundance of these insects (Gómez and Zamora 1992). However, ant pollination syndrome is related to dry habitats (Hickman 1974, Wyatt 1981) the habitat of M. oligantha presented an abundance of water.

The anthesis of M. oligantha species starts in cold early morning and raining days during blooming were common, under these conditions, instead of ants, flies are the main visitors, and when the weather was hotter the ants visit the flowers (Farinasso, personal observations). The importance of flies in ‘adverse’ conditions was already noticed by Faegri and van der Pijl (1979).

From one season to another, the relative frequency of flies to ants switched in M. oligantha.

Traits related to fly pollination (myophily) do not possess phenotypic specialization (sensu Ollerton et al. 2007) that might difficult this syndrome characterization, such as small radial flowers, small nectar quantity and variable concentration, diurnal anthesis and easy access to nectar (Faegri and van der Pijl 1979, Willmer 2011), and all these traits were present in M. oligantha. Normally, myophily flowers are visited by other insects orders (Endress 1994), but the flies can represent most frequency and act as main pollinators (Pombal and Morellato 1995, Niemirski and Zych 2011, Zych et al.

2013). Generalist plants can adapt to several pollinators groups, and the highest relative frequency of each group is related to its selection power on floral traits (Sahli and Conner 2011). Even if two

74 groups of pollinators present very different effectiveness, the plant is a generalist and can be adapted to these available pollinators (Aigner 2001). It seems that M. oligantha featured traits that are suitable for two different and unspecialized groups of pollinators, the ants and flies.

It is assumed that generalist systems are unlikely to fit in pollination syndromes or specialization (Herrera 1989, Gómez and Zamora 1999, Thompson and Wilson 2008, Gómez et al.

2014a). However, some were able to qualify an intermediate floral phenotype, pollinated and adapted to completely different functional groups and, despite that, still is characterized as generalist system

(Aigner 2005, Martén-Rodríguez et al. 2009, 2010, Wilson et al. 2017). It was shown that in generalist plants, abundant pollinators might exert strong selection on floral traits, even the low efficient ones (Gómez et al. 2014b). While M. oligantha presents a floral phenotype that appeared to be adapted to two main functional groups with similar importance, M. violacea despite being pollinated by several smaller groups tends to be ecologically specialized in bees. Therefore, our findings support the pollination systems of these two species of Manihot as a moderate generalist.

Pollinator filtering in a generalized monoecious system

We suggested that monoecy can function as a filter in the Manihot through female flowers.

Both species have functional groups that do not visit female flowers, whereas male flowers are always visited. What seems to determine this filtering is both floral display sex ratio and visitor relative abundance. The floral display of the two species of Manihot is male-biased with different degrees. Moreover, female display (and not male) in generalist plants was showed to be under a selection of pollinators (Sahli and Conner 2011). The sex ratio of M. violacea is male-biased and many functional groups with relatively low abundance are not recorded in female flowers.

Differently, the sex ratio of floral display in M. oligantha is more similar. It is known that increasing female sex ratio can promote a differential response on pollinators, making female flowers being more visited, and even more similar sex ratio could promote alike visitation (Yvonne and Wardle

2007). In M. oligantha, wasps and short-tongued bees visited both floral types even though their relative abundance was very low and only the lowest abundant functional group formed by small beetles did not visit female flowers. Likely due to the balanced sex ratio display, the filtering is more subtle in this latter species.

Our study indicates that the main pollinator groups do not discriminate against female flowers. Actually, nectar traits revealed that female flowers tend to higher volume or sugar concentration and might explain why pollinators visit this floral type more than expected. For the bees, these results are conflicting as this group frequently discriminates against female flowers. In addition to, other studies points out that flies and wasps prefer male flowers (Bell et al. 1984,

Ashman 2000). However, the female flower had lower nectar volume in the case of these latter authors. In studies that female flowers had higher nectar volume, this floral type was more attractive to pollinators including bees and flies (Gonzalez et al. 1995, Cervantes et al. 2017). Although in

Manihot female flowers are smaller in length and pollinators often prefer larger flowers (Conner and

Rush 1996, Kawagoe and Suzuki 2003, Parachnowitsch and Kessler 2010), the diameter is similar between sexes, making the size perhaps equally attractive. If female flowers are more attractive to the most important pollinators, what guarantees that they visit male flowers? The number of female flowers could maintain system stability.

Apis mellifera was shown to visits M. esculenta (Kawano 1980, Nassar and Carvalho 1990) and this bee is recorded in our field excursions, however, it does not visit the Manihot species.

Additionally, the richness of bee species visiting the Manihot flowers is low when compared to the

Cerrado fauna of the group (Pacheco Filho et al. 2015). Unfortunately, we could not analyze the chemical composition of nectar, but considering that Manihot has many secondary compounds

(reviewed in Blagbrough et al. 2010) the taste of nectar is probably toxic at some extent. Some compounds such as alkaloids might repel certain floral visitors (Kessler and Baldwin 2007).

Alkaloids levels in leaves could be relatively correlated with alkaloids levels in nectar (Adler et al.

2006). It is known that M. esculenta have prominent alkaloids contents in its leaves (Ebuehi et al.

2006), likely nectar contains alkaloids, which might be higher in the studied wild species and thus explains the fact that many bees did not visit the flowers. Besides, pollen collecting was uncommon, suggesting unpalatability. Consequently, the nectar and pollen traits could also act in filtering.

Furthermore, sugar concentration is especially low in male flowers of M. oligantha, below the concentration that is suitable for bees (Roubik et al. 1995, Biesmeijer et al. 1999, Cnaani et al.

2006). Indeed, M. oligantha barely received visits from bees. Male flowers of this species also featured low nectar volume that consisted with the lower nectary measurements on this floral type.

The nectar volume could be influenced by the predominant pollinators and play a big role in driving visitation (Wolff 2006). Little nectar volume might be unattractive and constraint visits by certain pollinators (Heinrich and Raven 1972, Hickman 1974, Willmer and Stone 2004). In contrast to this species, M. violacea present nectar with equal volume among floral types and the bees had the greatest importance on pollination. Floral rewards and attraction structures are usually gender-biased

(Wilson et al. 1994) and nectar rewarding plants with unisexual flowers present similar frequency of biased-nectar for each sex (Willson and Agren 1989, Delph 1996). Depending on which sex function is limited (pollen donation or receipt), it will be favored, i.e. nectar rewarding taxa that are bee- pollinated, the female is frequently the most rewarding flower, whereas in taxa bird-pollinated had generally male-biased nectar (Carlson and Harms 2006). If female flowers are common, attractive traits tent to be female-biased (Ashman and Diefenderfer 2001). In fact, female flowers are abundant in M. oligantha floral display and the nectar trait presents markedly sexual dimorphism. Therefore, we suppose that a more attractive female flower is being selected in M. oligantha.

In plants with unisexual flowers, as observed in M. violacea, it is expected and common the male-biased floral display, due to the less costly male function per flower compared to the female function (Lloyd and Webb 1977, Delph 1996, Costich and Meagher 2001, Huang et al. 2006).

However, M. oligantha have similar sex ratio in the floral display. The selection on the female

77 component in M. oligantha can be seen in other floral traits besides the nectar and floral display: in anthesis. The greater investment of this latter species is seen in floral longevity that counts with 11 hours of duration, in comparison with six hours in M. violacea. Long floral anthesis might be related to greater efforts to ensure successful pollination (Primack 1985, Ashman and Schoen 1994, Rathcke

2003). If the floral longevity is longer, the species might restrict less the flowers, which seems the case in M. oligantha. In contrast, M. violacea anthesis schedule is unusual that nectar is offered in the first hottest hours of the day and is readily explored by Partamona. It is known that large bees are easier to overheating than small bees, during foraging in full sunlight (Hrncir and Maia-Silva). We suggest that anthesis is restrictive in M. violacea and contribute to ecological specialization in this species.

In summary, the monoecy does act as pollinator filtering through female component. The apparent unspecialized flower morphology and the common nectar rewards can lead us to expect several visitors and pollinators. Indeed, we detect many functional groups of potential pollinators in the Manihot species, but there are species-specific filters, such as nectar traits and floral anthesis.

Our findings reinforce that even generalist systems have floral features that restricts pollinators assemblage.

REFERENCES

Abrahamczyk, S., Lozada-gobilard, S., Ackermann, M., Fischer, E., Krieger, V., Redling, A., & Weigend, M. (2017). A question of data quality — Testing pollination syndromes in Balsaminaceae. PLoS ONE, 1–14.

Adler, L. S., Wink, M., Distl, M., & Lentz, A. J. (2006). Leaf herbivory and nutrients increase nectar alkaloids. Ecology Letters, 9, 960–967. https://doi.org/10.1111/j.1461-0248.2006.00944.x

Aigner, P. A. (2001). Optimality modeling and fitness trade-offs: when should plants become pollinator specialists? Oikos, 95(1), 177–184.

Aigner, P. A. (2005). Variation in pollination performance gradients in a Dudleya species complex: Can generalization promote floral divergence? Functional Ecology, 19(4), 681–689. https://doi.org/10.1111/j.1365-2435.2005.01009.x

Armbruster, W. S. (2014). Floral specialization and angiosperm diversity: Phenotypic divergence, fitness trade-offs and realized pollination accuracy. AoB PLANTS, 6, 1–24. https://doi.org/10.1093/aobpla/plu003

Armbruster, W. S. (2017). The specialization continuum in pollination systems: diversity of concepts and implications for ecology, evolution and conservation. Functional Ecology, 31(1), 88–100. https://doi.org/10.1111/1365-2435.12783

Ashman, T. L. (2000). Pollinator selectivity and its implications for the evolution of dioecy and sexual dimorphism. Ecology, 81(9), 2577–2591.

Ashman, T. L., & Diefenderfer, C. (2001). Sex ratio represents a unique context for selection on attractive traits: consequences for the evolution of sexual dimorphism. The American Naturalist, 157(3), 334–347. https://doi.org/10.1086/319192

Ashman, T. L., & Schoen, D. J. (1994). How long should flowers live? Nature, 371, 788–791.

Ashworth, L., Aguilar, R., & Mart, S. (2015). Pollination Syndromes: A Global Pattern of Convergent Evolution Driven by the Most Effective Pollinator. In: Pontarotti P. (Eds) Evolutionary Biology: Biodiversification from Genotype to Phenotype. Springer, Cham, (July). https://doi.org/10.1007/978-3-319-19932-0

Bawa, K. S. (1977). The reproductive biology of Cupania guatemalensis Radlk. (Sapindaceae). Evolution, 31, 52–63.

Bawa, S. H. (1980). Evolution of dioecy in flowering plants. Annual Review of Ecology and Systematics, 11, 15–39.

Bell, G., Lefebvre, L., Giraldeau, L. A., & Weary, D. (1984). Partial preference of insects for the male flowers of an annual herb. Oecologia, 64(3), 287–294. https://doi.org/10.1007/BF00379123

Biesmeijer, J. C., & Richter, J. a P. (1999). Niche differentiation in nectar-collecting stingless bees: the infuence of morphology, oral choice and interference competition. Ecological Entomology, 24(4), 380–388. https://doi.org/10.1046/j.1365-2311.1999.00220.x

Brosi, B. J. (2016). Pollinator specialization: from the individual to the community. New Phytologist, 210, 1190–1194.

Burger, H., Jürgens, A., Ayasse, M., & Johnson, S. D. (2017). Floral signals and filters in a wasp- and a bee-pollinated Gomphocarpus species (Apocynaceae: Asclepiadeae). Flora, 232, 83–91. https://doi.org/10.1016/j.flora.2017.05.009

Carlson, J. E., & Harms, K. E. (2006). The evolution of gender-biased nectar production in hermaphroditic plants. The Botanical Review, 72(2), 179–205. https://doi.org/10.1663/0006- 8101(2006)72[179:TEOGNP]2.0.CO;2

Cervantes, C., Alvarez, A., & Cuevas, E. (2017). Small but attractive: female biased nectar production and floral visitors in a dimorphic shrub. Plant Biology, 20, 160–164. https://doi.org/10.1111/ijlh.12426

Charlesworth, D. (1993). Why are unisexual flowers associated with wind pollination and unspecialized pollinators? The American Naturalist, 141(3), 481–490.

Cnaani, J., Thomson, J. D., & Papaj, D. R. (2006). Flower choice and learning in foraging bumblebees: Effects of variation in nectar volume and concentration. Ethology, 112(3), 278– 285. https://doi.org/10.1111/j.1439-0310.2006.01174.x

Conner, J. K., & Rush, S. (1996). Effects of flower size and number on pollinator visitation to wild radish, Raphanus raphanistrum. Oecologia, 105, 509–516. https://doi.org/10.1007/BF00330014

Davila, Y. C., & Wardle, G. M. (2007). Bee boys and fly girls: do pollinators prefer male or female umbels in protandrous parsnip, Trachymene incisa (Apiaceae)? Austral Ecology, 32, 798–807. https://doi.org/10.1111/j.1442-9993.2007.01757.x de Vega, C. De, Ortiz, P. L., Herrera, C. M., & Talavera, S. (2009). The ant-pollination system of Cytinus hypocistis (Cytinaceae), a Mediterranean root holoparasite. Annals of Botany, 103, 1065–1075. https://doi.org/10.1093/aob/mcp049 de Vega, C., & Gómez, J. M. (2014). Polinización por hormigas: conceptos, evidencias y futuras direcciones. Ecossistemas, 23(3), 48–57.

Del-Claro, K., Rodriguez-Morales, D., Calixto, E. S., Martins, A. S., & Torezan-Silingardi, H. M. (2019). Ant pollination of Paepalanthus lundii (Eriocaulaceae) in Brazilian savanna. Annals of Botany, 1–7. https://doi.org/10.1093/aob/mcz021

Delph L. F. (1996). Flower size dimorphism in plants with unisexual flowers. In: Lloyd DG, Barrett SCH, eds. Floral biology: studies on floral evolution in animal-pollinated plants. New York, NY, USA: Chapman & Hall, 217– 237.

Delph, L. F., Galloway, L. F., & Stanton, M. L. (1996). Sexual dimorphism in flower size. The American Naturalist, 148(2), 299–320. https://doi.org/10.1086/285926

Dodonov, P., Xavier, R. D. O., Tiberio, F. C. S., Lucena, I. C. De, Zanelli, C. B., & Matos, D. M. S. (2014). Driving factors of small-scale variability in a savanna plant population after a fire. Acta Oecologica, 56, 47–55. https://doi.org/10.1016/j.actao.2014.02.003

Domingos-Melo, A., Nadia, T. D. L., & Machado, I. C. (2017). Complex flowers and rare pollinators: does ant pollination in Ditassa show a stable system in Asclepiadoideae (Apocynaceae)? Arthropod-Plant Interactions. https://doi.org/10.1007/s11829-017-9499-3

Ebuehi, O. A. T., Babalola, O., Ahmed, Z. 2005. Phytochemical, nutritive and anti-nutritive composition of cassava (Manihot esculenta L) tubers and leaves. Nigerian Food Journal Vol. 23: 40-46

Endress PK. (1994). Diversity and Evolutionary Biology of Tropical Flowers. Cambridge University Press, Cambridge, UK.

Etten, M. L. Van, & Chang, S. (2014). Frequency-dependent pollinator discrimination acts against female plants in the gynodioecious Geranium maculatum. Annals of Botany, 114, 1769–1778. https://doi.org/10.1093/aob/mcu204

Faegri, K., & van der Pijl, L. (1979). The principles of pollination ecology. 3rd ed. Oxford: Pergamon Press.

Flaig, I. C., Aguilar, I., Schmitt, T., & Jarau, S. (2016). An unusual recruitment strategy in a mass- recruiting stingless bee, Partamona orizabaensis. Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology, 202(9–10), 679–690. https://doi.org/10.1007/s00359-016-1111-2

Gómez, J. M., & Zamora, R. (1992). Pollination by ants: consequences of the quantitative effects on a mutualistic system. Oecologia, 410–418.

Gómez, J. M., & Zamora, R. (1999). Generalization vs. specialization in the pollination system of Hormathophylla spinosa (Cruciferae). Ecology, 80(3), 796–805.

Gómez, J. M., Muñoz-Pajares, A. J., Abdelaziz, M., Lorite, J., & Perfectti, F. (2014a). Evolution of pollination niches and floral divergence in the generalist plant Erysimum mediohispanicum. Annals of Botany, 113(2), 237–249. https://doi.org/10.1093/aob/mct186

Gómez, J. M., Perfectti, F., & Klingenberg, C. P. (2014b). The role of pollinator diversity in the evolution of corolla-shape integration in a pollination-generalist plant clade. Philosophical Transactions of the Royal Society B, 369(1649). https://doi.org/10.1098/rstb.2013.0257

Gong, Y., & Huang, S. (2009). Floral symmetry: pollinator-mediated stabilizing selection on flower size in bilateral species. Proceedings of the Royal Society B, 276, 4013–4020. https://doi.org/10.1098/rspb.2009.1254

Gonzalez, A., Rowe, C. L., Weeks, P. J., Whittle, D., Gilbert, F. S., & Barnard, C. J. (1995). Flower choice by honey bees (Apis mellifera L.): sex-phase of flowers and preferences among nectar and pollen foragers. Oecologia, 101(2), 258–264. https://doi.org/10.1007/BF00317292

Heinrich, B., & Raven, P. H. (1972). Energetics and Pollination Ecology. Science, 176, 597–602.

Herrera, C. M. (1988). Variation in mutualisms: the spatio-temporal mosaic of a pollinator assemblage. Biological Journal of the Linnean Society, 35, 95–125.

Herrera, C. M. (1989). Pollinator abundance, morphology, and flower visitation rate: analysis of the “quantity” component in a plant-pollinator system. Oecologia, 80, 241–248.

Herrera, C. M., Canto, A., Pozo, M. I., & Bazaga, P. (2010). Inhospitable sweetness: nectar filtering of pollinator-borne inocula leads to impoverished, phylogenetically clustered yeast communities. Proceedings of the Royal Society B, 277, 747–754. https://doi.org/10.1098/rspb.2009.1485

Hickman, J. C. (1974). Pollination by Ants: A Low-Energy System. Science, 184, 1290–1292.

Hrncir, M., & Maia-silva, C. (2013). The fast versus the furious – On competition, morphological foraging traits, and foraging in stingless bees. Stingless bees process honey and pollen in cerumen pots. Facultad de Farmacia y Bioanálisis, Universidad de Los Andes, Mérida, Venezuela.

Ibarra-isassi, J., & Sendoya, S. F. (2016). Ants as floral visitors of Blutaparon portulacoides (A. St- Hil.) Mears (Amaranthaceae): an ant pollination system in the Atlantic Rainforest. Arthropod- Plant Interactions. https://doi.org/10.1007/s11829-016-9429-9

John Fox and Sanford Weisberg (2011). An {R} Companion to Applied Regression, Second Edition. Thousand Oaks CA: Sage. URL: http://socserv.socsci.mcmaster.ca/jfox/Books/Companion

John Fox, Sanford Weisberg and Brad Price (2018). carData: Companion to Applied Regression Data Sets. R package version 3.0-1. https://CRAN.R-project.org/package=carData

Johnson, S. D., & Steiner, K. E. (2000). Generalization versus specialization in plant pollination systems. Trends in Ecology and Evolution, 15(4), 140–143. https://doi.org/10.1016/S0169- 5347(99)01811-X

Johnson, S. D., Hargreaves, A., & Brown, M. (2006). Dark, bitter-tasting nectar functions as a filter of flower visitors in a bird-pollinated plant. Ecology, 87(11), 2709–2716.

Joly, S., Lambert, F., Alexandre, H., Clavel, J., Léveillé-Bourret, É., & Clark, J. L. (2018). Greater pollination generalization is not associated with reduced constraints on corolla shape in Antillean plants. Evolution, 72(2), 244–260. https://doi.org/10.1111/evo.13410

Kawagoe, T., & Suzuki, N. (2003). Flower-size dimorphism avoids geitonogamous pollination in a nectarless monoecious plant Akebia quinata. International Journal of Plant Sciences, 164(6), 893–897.

Kawano, K. 1980. Cassava: In hybridization of Crop plants, Fehr, W. R. and Hadley, H. H. (eds.), ASA/CSSA, Madison.

Lázaro, A., Hegland, S. J., & Totland, O. (2008). The relationships between floral traits and specificity of pollination systems in three Scandinavian plant communities. Oecologia, 157, 249–257. https://doi.org/10.1007/s00442-008-1066-2

Li, J., & Huang, S. (2009). Effective pollinators of Asian sacred lotus (Nelumbo nucifera): Contemporary pollinators may not reflect the historical pollination syndrome. Annals of Botany, 104, 845–851. https://doi.org/10.1093/aob/mcp173

Lloyd D. G., Webb C. J. (1977). Secondary sex characters in plants. Botanical Review 43: 177–216.

Martén-Rodríguez, S., Almarales-Castro, A., & Fenster, C. B. (2009). Evaluation of pollination syndromes in Antillean Gesneriaceae: evidence for bat, hummingbird and generalized flower. Journal of Ecology, 97, 348–359. https://doi.org/10.1111/j.1365-2745.2008.01465.x

Martén-Rodríguez, S., Fenster, C. B., Agnarsson, I., Skog, L. E., Zimmer, E. A., & Marte, S. (2010). Evolutionary breakdown of pollination specialization in a Caribbean plant radiation. New Phytologist, 403–417.

Mitchell, R. J. (2004). Heritability of nectar traits: why do we know so little? Ecology, 85(6), 1527– 1533.

Nassar, N. M. A., Mendoza, F. J. M., & Sano, N. (2012). Some interesting cassava cultivars: 9 – ICB 300. Geneconserve, 11(43), 22–25.

Neal, P. R., Dafni, A., & Giurfa, M. (1998). Floral Symmetry and Its Role in Plant-Pollinator Systems: Terminology, Distribution, and Hypotheses. Annual Review of Ecology and Systematics, 29, 345–373.

Niemirski, R., & Zych, M. (2011). Fly pollination of dichogamous Angelica sylvestris (Apiaceae): How (functionally) specialized can a (morphologically) generalized plant be? Plant Systematics and Evolution. https://doi.org/10.1007/s00606-011-0454-y

Olesen, J. M., Dupont, Y. L., Ehlers, B. K., & Hansen, D. M. (2007). The openness of a flower and its number of flower-visitor species. Taxon, 56(3), 729–736. https://doi.org/10.2307/25065857

Ollerton, J., Alarcon, R., Waser, N. M., Price, M. V., Watts, S., Cranmer, L., Rotenberry, J. (2009). A global test of the pollination syndrome hypothesis. Annals of Botany, 103, 1471–1480. https://doi.org/10.1093/aob/mcp031

Ollerton, J., Killick, A., Lamborn, E., Watts, S., & Whiston, M. (2007). Multiple Meanings and Modes: On the Many Ways to Be a Generalist Flower. Taxon, 56(3), 717–728. https://doi.org/10.2307/25065856

Ollerton, J., Rech, A. R., Waser, N. M., & Price, M. V. (2015). Using the literature to test pollination syndromes — some methodological cautions. Journal of Pollination Ecology, 16(17), 119–125.

Pacheco Filho, A. J. S., Verola, C. F., Lima Verde, L. W., & Freitas, B. M. (2015). Bee-flower association in the Neotropics: implications to bee conservation and plant pollination. Apidologie, 46, 530–541. https://doi.org/10.1007/s13592-014-0344-8

Padyšáková, E., Bartoš, M., Tropek, R., & Janeček, Š. (2013). Generalization versus specialization in pollination systems: visitors, thieves, and pollinators of Hypoestes aristata (Acanthaceae). PLoS ONE, 8(4), 1–8. https://doi.org/10.1371/journal.pone.0059299

Parachnowitsch, A. L., & Kessler, A. (2010). Pollinators exert natural selection on flower size and floral display in Penstemon digitalis. New Phytologist, 188, 393–402.

Perfectti, F., Gómez, J. M., & Bosch, J. (2009). The functional consequences of diversity in plant pollinator- interactions. Oikos, 1430–1440. https://doi.org/10.1111/j.1600-0706.2009.17491.x

Petanidou, T., Kallimanis, A. S., Tzanopoulos, J., Sgardelis, S. P., & Pantis, J. D. (2008). Long-term observation of a pollination network: fluctuation in species and interactions, relative invariance of network structure and implications for estimates of specialization. Ecology Letters, 11, 564– 575. https://doi.org/10.1111/j.1461-0248.2008.01170.x

Pombal, E.C.P. & P.C Morellato. 1995. Polinização por moscas em Dendropanax cuneatum Decne & Planch (Araliaceae) em floresta semidecídua no Sudeste do Brasil. Revista Brasileira de Botânica, 18(2), 157-162.

Primack, R. B. (1985). Longevity of individual flowers. Annual Review of Ecology and Systematics, 16, 15–37.

R Core Team (2018). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.

Rathcke, B. J. (2003). Floral longevity and reproductive assurance: seasonal patterns and an experimental test with Kalmia latifolia (Ericaceae). American Journal of Botany, 90(9), 1328– 1332. https://doi.org/10.3732/ajb.90.9.1328

Reverté, S., Retana, J., Valle, C., & Valle, C. (2016). Pollinators show flower colour preferences but flowers with similar colours do not attract similar pollinators. Annals of Botany, 118, 249–257. https://doi.org/10.1093/aob/mcw103

Rogers, D. J., & Appan, S. G. (1973). Manihot Manihotoides (Euphorbiaceae). Flora Neotropica, 13, 1–272.

Rosas-Guerrero, V., Aguilar, R., Martén-Rodríguez, S., Ashworth, L., Lopezaraiza-Mikel, M., Bastida, J. M., & Quesada, M. (2014). A quantitative review of pollination syndromes: Do floral traits predict effective pollinators? Ecology Letters, 17(3), 388–400. https://doi.org/10.1111/ele.12224

Rostás, M., Bollmann, F., Saville, D., & Riedel, M. (2018). Ants contribute to pollination but not to reproduction in a rare calcareous grassland forb. PeerJ, 1–16. https://doi.org/10.7717/peerj.4369

Roubik, D. W., Yanega, D., Aluja, S., Buchmann, S. L., & Inouye, D. W. (1995). On optimal nectar foraging by some tropical bees (Hymenoptera: Apidae). Apidologie, 26(3), 197–211. https://doi.org/10.1051/apido:19950303

Russell Lenth (2018). Emmeans: Estimated Marginal Means, aka Least-Squares Means. R package version 1.2.4. https://CRAN.R-project.org/package=emmeans

Sahli, H. F., & Conner, J. K. (2006). Characterizing ecological generalization in plant-pollination systems. Oecologia, 365–372. https://doi.org/10.1007/s00442-006-0396-1

Sahli, H. F., & Conner, J. K. (2011). Testing for conflicting and nonadditive selection: Floral adaptation to multiple pollinators through male and female fitness. Evolution, 65(5), 1457– 1473. https://doi.org/10.1111/j.1558-5646.2011.01229.x

Sargent, R. D. (2004). Floral symmetry affects speciation rates in angiosperms. Proceedings of the Royal Society B, 271, 603–608. https://doi.org/10.1098/rspb.2003.2644

Schindelin, J., Arganda-Carreras, I. & Frise, E. et al. (2012). Fiji: an open-source platform for biological-image analysis. Nature methods 9(7), 676-682, PMID 22743772, doi:10.1038/nmeth.2019

Shuttleworth, A., & Johnson, S. D. (2010). The missing stink: Sulphur compounds can mediate a shift between fly and wasp pollination systems. Proceedings of the Royal Society B: Biological Sciences, 277(1695), 2811–2819. https://doi.org/10.1098/rspb.2010.0491

Silva, J. F., Farinas, M. R., Felfili, J. M., & Klink, C. A. (2006). Spatial heterogeneity , land use and conservation in the cerrado region of. Journal of Biogeography, 33, 536–548. https://doi.org/10.1111/j.1365-2699.2005.01422.x

Simon, M. F., Reis, T. S., Mendoza F., J. M., Arquelão, T. K. M., Bringel, J. B. A., Noronha, S. E., Scariot, A. (2018). Conservation assessment of cassava wild relatives in central Brazil.

Biodiversity and Conservation, (0123456789), 1–24. https://doi.org/10.1007/s10531-018-1626- 7

Thomson, J. D., & Wilson, P. (2008). Explaining evolutionary shifts between bee and hummingbird pollination: convergence, divergence, and directionality. International Journal of Plant Sciences, 169(1), 23–38. https://doi.org/10.1086/523361

Totland, Ø. (2004). No evidence for a role of pollinator discrimination in causing selection on flower size through female reproduction. Oikos, 3, 558–564.

Vázquez, D. P., Morris, W. F., & Jordano, P. (2005). Interaction frequency as a surrogate for the total effect of animal mutualists on plants. Ecology Letters, 8, 1088–1094. https://doi.org/10.1111/j.1461-0248.2005.00810.x

Vieira, L. D. J., Soares, T. L., Rossi, M. L., Alves, A. C., dos Santos, F. A. R., & Souza, F. V. D. (2012). Viability, production and morphology of pollen grains for diff erent species in the genus Manihot (Euphorbiaceae). Acta Botanica Brasilica, 26(2), 350–356.

Vincent, H., Wiersema, J., Kell, S., Fielder, H., Dobbie, S., Castañeda-Álvarez, N. P., Maxted, N. (2013). A prioritized crop wild relative inventory to help underpin global food security. Biological Conservation, 167, 265–275. https://doi.org/10.1016/j.biocon.2013.08.011

Wang, S., Fu, W. L., Du, W., Zhang, Q., Li, Y., Lyu, Y. S., & Wang, X. F. (2018). Nectary tracks as pollinator manipulators: The pollination ecology of Swertia bimaculata (Gentianaceae). Ecology and Evolution, 1–21. https://doi.org/10.1002/ece3.3838

Waser, N. M., Chittka, L., Price, M. V, Williams, N. M., & Ollerton, J. (1996). Generalization in Pollination Systems, and Why it Matters. Ecology, 77(4), 1043–1060. https://doi.org/10.2307/2265575

Wickham H. (2016). ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York. ISBN 978-3-319-24277-4, http://ggplot2.org.

Willmer, P. 2011. Pollination and floral ecology. Princeton University Press, Princeton.

Willmer, P. G., & Stone, G. N. (2004). Behavioral, ecological, and physiological determinants of the activity patterns of bees. Advances in the Study of Behavior, 34, 347–466.

Willson, M. F., & Ågren, J. (1989). Differential floral rewards and pollination by deceit in unisexual flowers. Oikos, 55(1), 23–29. https://doi.org/10.2307/3565868

Wilson P, Castellanos MC, Hogue JN, Thomson JD, Armbruster WS. 2004. A multivariate search for pollination syndromes among penstemons. Oikos 104: 345–361

Wilson, P., Thomson, J. D., Stanton, M. L., & Rigney, L. P. (1994). Beyond floral batemania: gender biases in selection for pollination success. The American Naturalist, 143(2), 283–296.

Wilson, T. C., Conn, B. J., & Henwood, M. J. (2017). Great Expectations: Correlations between Pollinator Assemblages and Floral Characters in Lamiaceae. International Journal of Plant Sciences, 178(3), 170–187. https://doi.org/10.1086/690023

Wolff, D. (2006). Nectar sugar composition and volumes of 47 species of Gentianales from a southern Ecuadorian montane forest. Annals of Botany, 97(5), 767–777. https://doi.org/10.1093/aob/mcl033

Wyatt, R. (1981). Ant-pollination of the granite outcrop endemic Diamorpha smallii (Crassulaceae). American Journal of Botany, 68(9), 1212–1217. https://doi.org/10.1002/j.1537- 2197.1981.tb07827.x

Yakimowski, S. B., Glaettli, M., & Barrett, S. C. H. (2011). Floral dimorphism in plant populations with combined versus separate sexes. Annals of Botany, 108(4), 765–776. https://doi.org/10.1093/aob/mcr025

Zapata, T. R., & Arroyo, M. T. K. (1978). Plant Reproductive Ecology of a Secondary Deciduous Tropical Forest in Venezuela. Biotropica, 10(3), 221–230.

Zych, M., Junker, R. R., Nepi, M., Stpiczy, M., Stolarska, B., & Roguz, K. (2018). Spatiotemporal variation in the pollination systems of a supergeneralist plant: is Angelica sylvestris (Apiaceae) locally adapted to its most effective pollinators? Annals of Botany, (August), 1–14. https://doi.org/10.1093/aob/mcy140

Zych, M., Michalska, B., & Krasicka-Korczynska, E. (2013). Myophily in a critically endangered umbelliferous plant Ostericum palustre Myophily in the critically endangered umbelliferous plant Ostericum palustre Besser (Apiaceae). Plant Systematics and Evolution. https://doi.org/10.1007/s00606-013-0870-2

APPENDIX

Floral visitors collected in Manihot species of the Cerrado

DZUB Floral Plant species Coordinate Date Order Family Genus Epithet Author number types -16.737399, - Latreille, 060958 M. oligantha male 3/i/2018 Coleoptera Megalopodidae sp. 47.689693 1802 -16.737399, - Duda, 060971 M. oligantha female 30/xii/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060963 M. oligantha female 30/xii/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060978 M. oligantha female 30/xii/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060983 M. oligantha female 2/i/2018 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060982 M. oligantha male 15/xii/2018 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060964 M. oligantha male 25/xi/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060965 M. oligantha male 3/i/2018 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060966 M. oligantha male 24/xi/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060967 M. oligantha male 25/xi/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060974 M. oligantha male 24/xi/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060969 M. oligantha male 24/xi/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060973 M. oligantha male 25/xi/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060975 M. oligantha male 25/xi/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060979 M. oligantha male 2/i/2018 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060977 M. oligantha male 2/i/2018 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060976 M. oligantha male 25/xi/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060984 M. oligantha male 2/i/2018 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060981 M. oligantha male 24/xi/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060972 M. oligantha male 31/xii/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Duda, 060980 M. oligantha male 31/xii/2017 Diptera Chloropidae Liohippelates sp. 47.689693 1929 -16.737399, - Loew, 060968 M. oligantha male 25/xi/2017 Diptera Lonchaeidae sp. 47.689693 1861 (Shannon -16.737399, - & Del 219796 M. oligantha female 15/xii/2018 Diptera Muscidae Sarcopromusca pruna 47.689693 Ponte, 1926) -16.737399, - Linnaeus, 060970 M. oligantha female 2/i/2018 Diptera Muscidae Musca domestica 47.689693 1758 -16.737399, - Linnaeus, 219797 M. oligantha female 15/xii/2018 Diptera Empididae Empis sp. 47.689693 1758 -16.737399, - Linnaeus, 219823 M. oligantha female 15/xii/2018 Diptera Empididae Empis sp. 47.689693 1758 -16.737399, - Linnaeus, 219801 M. oligantha male 15/xii/2018 Diptera Empididae Empis sp. 47.689693 1758

-16.737399, - Mayr, 219824 M. oligantha female 15/xii/2018 Hymenoptera Formicidae Camponotus crassus 47.689693 1862 -16.737399, - Mayr, 219799 M. oligantha male 15/xii/2018 Hymenoptera Formicidae Camponotus crassus 47.689693 1862 -16.737399, - Emery, 219800 M. oligantha male 15/xii/2018 Hymenoptera Formicidae Camponotus melanoticus 47.689693 1894

- 16.737399, - Forel, 060954 M. oligantha female 2/i/2018 Hymenoptera Formicidae Dorymyrmex brunneus 47.689693 1908 -16.737399, - Forel, 060961 M. oligantha male 3/i/2018 Hymenoptera Formicidae Dorymyrmex brunneus 47.689693 1908 -16.737399, - Forel, 060957 M. oligantha male 31/xii/2017 Hymenoptera Formicidae Dorymyrmex brunneus 47.689693 1908 -16.737399, - Forel, 060956 M. oligantha male 2/i/2018 Hymenoptera Formicidae Dorymyrmex brunneus 47.689693 1908 -16.737399, - Smith, F., 219798 M. oligantha female 15/xii/2018 Hymenoptera Formicidae Ectatomma brunneum 47.689693 1858 -16.737399, - Smith, F., 219794 M. oligantha female 15/xii/2018 Hymenoptera Formicidae Ectatomma brunneum 47.689693 1858 -16.737399, - Smith, F., 219795 M. oligantha female 15/xii/2018 Hymenoptera Formicidae Ectatomma brunneum 47.689693 1858 -16.737399, - (Smith, F., 219802 M. oligantha male 15/xii/2018 Hymenoptera Formicidae Pseudomyrmex termitarius 47.689693 1855) -16.737399, - Forel, 060955 M. oligantha male 2/i/2018 Hymenoptera Formicidae Solenopsis tridens 47.689693 1911 -16.737399, - Robertson, 060962 M. oligantha male 3/i/2018 Hymenoptera Halictidae Dialictus sp. 47.689693 1902 -16.737399, - (Haliday 060959 M. oligantha female 3/i/2018 Hymenoptera Vespidae Polybia ignobilis 47.689693 1836) -15.796089,- Latreille, 060909 M. violacea female 09/xii/2017 Coleoptera Carabidae sp 47.805440 1802 -15.796089,- Latreille, 061012 M. violacea female 17/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 061007 M. violacea female 19/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 061011 M. violacea female 19/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060928 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060925 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060875 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060927 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060878 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060926 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060899 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060904 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060902 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060903 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060901 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060900 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060897 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060898 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060905 M. violacea male 18/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 060988 M. violacea male 19/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 061006 M. violacea male 20/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 061003 M. violacea male 20/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 061004 M. violacea male 20/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 060940 M. violacea male -15.796089,- 22/i/2018 Coleoptera Chrysomelidae sp. 1 Latreille,

47.805440 1802 -15.796089,- Latreille, 061021 M. violacea male 22/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.796089,- Latreille, 061022 M. violacea male 22/i/2018 Coleoptera Chrysomelidae sp. 1 47.805440 1802 -15.717940, - Latreille, 060949 M. violacea male 29/i/2018 Coleoptera Chrysomelidae sp. 1 47.851572 1802

- 15.717940, - Latreille, 060995 M. violacea male 30/i/2018 Coleoptera Chrysomelidae sp. 1 47.851572 1802 -15.796089,- Latreille, 061016 M. violacea female 19/i/2018 Coleoptera Chrysomelidae sp. 2 47.805440 1802 -15.796089,- Latreille, 060929 M. violacea male 20/i/2018 Coleoptera Chrysomelidae sp. 2 47.805440 1802 -15.796089,- Latreille, 060919 M. violacea male 20/i/2018 Coleoptera Chrysomelidae sp. 2 47.805440 1802 -15.796089,- Latreille, 060856 M. violacea male 10/xii/2017 Coleoptera Curculionidae sp. 47.805440 1802 -15.796089,- Latreille, 060910 M. violacea male 09/xii/2017 Coleoptera Curculionidae sp. 47.805440 1802 -15.796089,- Latreille, 060834 M. violacea male 22/xi/2017 Hymenoptera Apidae Ceratina sp. 1 47.805440 1802 -15.796089,- Latreille, 060835 M. violacea male 22/xi/2017 Hymenoptera Apidae Ceratina sp. 1 47.805440 1802 -15.796089,- Latreille, 060837 M. violacea male 22/xi/2017 Hymenoptera Apidae Ceratina sp. 1 47.805440 1802 -15.796089,- Latreille, 060838 M. violacea male 22/xi/2017 Hymenoptera Apidae Ceratina sp. 2 47.805440 1802 -15.796089,- Latreille, 061020 M. violacea male 22/i/2018 Hymenoptera Apidae Ceratina sp. 2 47.805440 1802 -15.796089,- Latreille, 060985 M. violacea male 24/xi/2016 Hymenoptera Apidae Ceratina sp. 3 47.805440 1802 -15.796089,- Latreille, 060813 M. violacea male 21/xi/2017 Hymenoptera Apidae Ceratina sp. 3 47.805440 1802 -15.796089,- Latreille, 060836 M. violacea male 22/xi/2017 Hymenoptera Apidae Ceratina sp. 3 47.805440 1802 -15.796089,- Latreille, 061018 M. violacea male 22/i/2018 Hymenoptera Apidae Ceratina sp. 3 47.805440 1802 -15.796089,- Latreille, 060944 M. violacea male 28/i/2018 Hymenoptera Apidae Ceratina sp. 3 47.805440 1802 -15.717940, - Latreille, 060937 M. violacea male 29/i/2018 Hymenoptera Apidae Ceratina sp. 3 47.851572 1802 -15.796089,- (Smith, 061014 M. violacea female 17/i/2018 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 061010 M. violacea female 17/i/2018 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 061013 M. violacea female 17/i/2018 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 061001 M. violacea female 18/iii/2016 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060999 M. violacea female 26/iii/2016 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 028653 M. violacea female 28/i/2018 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060986 M. violacea male 24/xi/2016 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060987 M. violacea male 24/xi/2016 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060833 M. violacea male 22/xi/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060809 M. violacea male 22/xi/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060913 M. violacea male 02/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060911 M. violacea male 02/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060912 M. violacea male 02/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060915 M. violacea male 02/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060914 M. violacea male 02/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060918 M. violacea male 10/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060854 M. violacea male 10/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863)

-15.796089,- (Smith, 060917 M. violacea male 10/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060853 M. violacea male 10/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060916 M. violacea male 10/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863)

- 15.796089,- (Smith, 061019 M. violacea male 12/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060862 M. violacea male 12/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060859 M. violacea male 12/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060857 M. violacea male 12/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060858 M. violacea male 12/xii/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060883 M. violacea male 17/i/2018 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060881 M. violacea male 17/i/2018 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060880 M. violacea male 17/i/2018 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060906 M. violacea male 18/i/2018 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060877 M. violacea male 18/i/2018 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060840 M. violacea male 22/xi/2017 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.796089,- (Smith, 060996 M. violacea male 26/iii/2016 Hymenoptera Apidae Partamona cupira 47.805440 1863) -15.717940, - (Smith, 060935 M. violacea male 29/i/2018 Hymenoptera Apidae Partamona cupira 47.851572 1863) -15.717940, - (Smith, 060933 M. violacea male 29/i/2018 Hymenoptera Apidae Partamona cupira 47.851572 1863) -15.717940, - (Smith, 060930 M. violacea male 29/i/2018 Hymenoptera Apidae Partamona cupira 47.851572 1863) -15.717940, - (Smith, 060936 M. violacea male 29/i/2018 Hymenoptera Apidae Partamona cupira 47.851572 1863) -15.717940, - (Smith, 060932 M. violacea male 29/i/2018 Hymenoptera Apidae Partamona cupira 47.851572 1863) -15.717940, - (Smith, 060931 M. violacea male 29/i/2018 Hymenoptera Apidae Partamona cupira 47.851572 1863) -15.717940, - (Smith, 060950 M. violacea male 30/i/2018 Hymenoptera Apidae Partamona cupira 47.851572 1863) -15.717940, - (Smith, 060943 M. violacea male 30/i/2018 Hymenoptera Apidae Partamona cupira 47.851572 1863) -15.717940, - Schwarz, 060947 M. violacea male 29/i/2018 Hymenoptera Apidae Plebeia sp. 47.851572 1938 -15.796089,- (Fabricius, 060817 M. violacea male 21/xi/2017 Hymenoptera Apidae Trigona spinipes 47.805440 1793) -15.796089,- (Smith, F., 061008 M. violacea female 19/i/2018 Hymenoptera Formicidae Camponotus blandus 47.805440 1858) -15.796089,- Mayr, 061009 M. violacea female 19/i/2018 Hymenoptera Formicidae Camponotus crassus 47.805440 1862 -15.796089,- Mayr, 060994 M. violacea male 18/i/2018 Hymenoptera Formicidae Camponotus crassus 47.805440 1862 -15.796089,- Mayr, 060993 M. violacea male 18/i/2018 Hymenoptera Formicidae Camponotus crassus 47.805440 1862 -15.796089,- Mayr, 060921 M. violacea male 18/i/2018 Hymenoptera Formicidae Camponotus crassus 47.805440 1862 -15.796089,- Mayr, 060922 M. violacea male 18/i/2018 Hymenoptera Formicidae Camponotus crassus 47.805440 1862 -15.796089,- Mayr, 060992 M. violacea male 19/i/2018 Hymenoptera Formicidae Camponotus crassus 47.805440 1862 -15.796089,- Mayr, 060991 M. violacea male 19/i/2018 Hymenoptera Formicidae Camponotus crassus 47.805440 1862 -15.717940, - Mayr, 060946 M. violacea female 30/i/2018 Hymenoptera Formicidae Camponotus novogranadensis 47.851572 1870 -15.717940, - Mayr, 060942 M. violacea male 29/i/2018 Hymenoptera Formicidae Camponotus novogranadensis 47.851572 1870 -15.717940, - Mayr, 060941 M. violacea female 30/i/2018 Hymenoptera Formicidae Dorymyrmex sp. 47.851572 1866 -15.796089,- (Smith, F., 060876 M. violacea male 18/i/2018 Hymenoptera Formicidae Pseudomyrmex termitarius 47.805440 1855) 060814 M. violacea male -15.796089,- 21/xi/2017 Hymenoptera Halictidae Augochlora sp. 1 Smith,

47.805440 1853 -15.796089,- Smith, 060920 M. violacea male 20/i/2018 Hymenoptera Halictidae Augochlora sp. 2 47.805440 1853 -15.796089,- Smith, 060832 M. violacea male 09/xii/2017 Hymenoptera Halictidae Augochlora sp. 3 47.805440 1853

- 15.796089,- Cockerell, 060815 M. violacea female 21/xi/2017 Hymenoptera Halictidae Augochloropsis sp. 1 47.805440 1897 -15.796089,- Cockerell, 060811 M. violacea male 22/xi/2017 Hymenoptera Halictidae Augochloropsis sp. 1 47.805440 1897 -15.796089,- Cockerell, 060860 M. violacea male 12/xii/2017 Hymenoptera Halictidae Augochloropsis sp. 1 47.805440 1897 -15.717940, - Cockerell, 060938 M. violacea male 29/i/2018 Hymenoptera Halictidae Augochloropsis sp. 1 47.851572 1897 -15.796089,- Cockerell, 060884 M. violacea male 17/i/2018 Hymenoptera Halictidae Augochloropsis sp. 2 47.805440 1897 -15.796089,- Cockerell, 061015 M. violacea male 21/i/2018 Hymenoptera Halictidae Augochloropsis sp. 3 47.805440 1897 -15.796089,- Moure, 060818 M. violacea male 21/xi/2017 Hymenoptera Halictidae Paraxystoglossa sp. 47.805440 1941 -15.796089,- Moure, 060816 M. violacea male 21/xi/2017 Hymenoptera Halictidae Paraxystoglossa sp. 47.805440 1941 -15.796089,- Moure, 060810 M. violacea male 22/xi/2017 Hymenoptera Halictidae Paraxystoglossa sp. 47.805440 1941 -15.796089,- Moure, 060839 M. violacea male 22/xi/2017 Hymenoptera Halictidae Paraxystoglossa sp. 47.805440 1941 -15.796089,- Schrottky, 060879 M. violacea male 17/i/2018 Hymenoptera Halictidae Rhinocorynura sp. 47.805440 1909 -15.796089,- Olivier, 060882 M. violacea male 17/i/2018 Hymenoptera Vespidae Polybia occidentalis 47.805440 1791 -15.796089,- (Olivier, 060812 M. violacea female 21/xi/2017 Hymenoptera Vespidae Polybia sericea 47.805440 1791) -15.796089,- (Olivier, 061005 M. violacea female 17/i/2018 Hymenoptera Vespidae Polybia sericea 47.805440 1791) -15.796089,- (Olivier, 060831 M. violacea male 09/xii/2017 Hymenoptera Vespidae Polybia sericea 47.805440 1791) -15.796089,- (Olivier, 060945 M. violacea male 09/xii/2017 Hymenoptera Vespidae Polybia sericea 47.805440 1791) -15.796089,- (Olivier, 060952 M. violacea male 09/xii/2017 Hymenoptera Vespidae Polybia sericea 47.805440 1791) -15.796089,- (Olivier, 060855 M. violacea male 10/xii/2017 Hymenoptera Vespidae Polybia sericea 47.805440 1791) -15.796089,- (Olivier, 060861 M. violacea male 12/xii/2017 Hymenoptera Vespidae Polybia sericea 47.805440 1791) -15.796089,- (Olivier, 061017 M. violacea male 17/i/2018 Hymenoptera Vespidae Polybia sericea 47.805440 1791) -15.717940, - (Olivier, 060948 M. violacea male 30/i/2018 Hymenoptera Vespidae Polybia sericea 47.851572 1791) -15.717940, - Lepeletier, 060939 M. violacea male 29/i/2018 Hymenoptera Vespidae Polybia sp. 47.851572 1836 -15.796089,- Linnaeus, 060990 M. violacea male 19/i/2018 Lepidoptera sp. 47.805440 1758 -15.796089,- Latreille, 060923 M. violacea male 18/i/2018 Orthoptera sp. 47.805440 1793 -15.717940, - Latreille, 060934 M. violacea male 29/i/2018 Orthoptera sp. 47.851572 1793